Accepted Manuscript

Long term Follow-up of the Trial of Routine ANgioplasty and Stenting after Fibrinolysis
to Enhance Reperfusion in Acute Myocardial Infarction (TRANSFER AMI)

Yaron Arbel, MD, Dennis T. Ko, MD, MSc, Andrew T. Yan, MD, Warren J. Cantor,
MD, Akshay Bagai, MD, MHS, Maria Koh, MSc, Maria Eberg, MSc, Mary Tan,
MSc, David Fitchett, MD, Bjug Borgundvaag, MD, PhD, John Ducas, MD, Michael
Heffernan, MD, PhD, Laurie J. Morrison, MD, Anatoly Langer, MD, MSc, Vladimir
Dzavik, MD, Shamir R. Mehta, MD, Shaun G. Goodman, MD, MSc

PII: S0828-282X(18)30150-8
DOI: 10.1016/j.cjca.2018.02.005
Reference: CJCA 2739

To appear in: Canadian Journal of Cardiology

Received Date: 4 December 2017

Revised Date: 5 February 2018
Accepted Date: 6 February 2018

Please cite this article as: Arbel Y, Ko DT, Yan AT, Cantor WJ, Bagai A, Koh M, Eberg M, Tan M,
Fitchett D, Borgundvaag B, Ducas J, Heffernan M, Morrison LJ, Langer A, Dzavik V, Mehta SR,
Goodman SG, on behalf of the TRANSFER-AMI Trial Investigators, Long term Follow-up of the Trial
of Routine ANgioplasty and Stenting after Fibrinolysis to Enhance Reperfusion in Acute Myocardial
Infarction (TRANSFER AMI), Canadian Journal of Cardiology (2018), doi: 10.1016/j.cjca.2018.02.005.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Long term Follow-up of the Trial of Routine ANgioplasty and Stenting after Fibrinolysis to

Enhance Reperfusion in Acute Myocardial Infarction (TRANSFER AMI)

Yaron Arbel, MDa, Dennis T. Ko, MD, MSc b,c, Andrew T. Yan, MD d, Warren J. Cantor, MDe,

PT
Akshay Bagai, MD, MHSd, Maria Koh, MScc, Maria Eberg, MScc, Mary Tan, MScj David

RI
Fitchett, MDd, Bjug Borgundvaag, MD, PhDf, John Ducas, MDg, Michael Heffernan, MD, PhDh,

Laurie J. Morrison, MDi, Anatoly Langer, MD, MScj, Vladimir Dzavik, MDk, Shamir R. Mehta,

SC
MDl, Shaun G. Goodman, MD, MSc d,j, on behalf of the TRANSFER-AMI Trial Investigators

U
Affiliations
AN
a
Department of Cardiology, Tel Aviv Medical Center, Sackler School of Medicine, Tel Aviv
b
University, Tel Aviv, Israel ; Schulich Heart Centre, Division of Cardiology, Sunnybrook
M

Health Sciences Centre, University of Toronto, Ontario, Canada; c Institute for Clinical

Evaluative Sciences (ICES), Toronto, Ontario, Canada; d St. Michael's Hospital, University of
D

e
Toronto, Toronto, Ontario, Canada; Department of Cardiology, Southlake Regional Health
TE

Centre, University of Toronto, Ontario, Canada; f Schwartz/Reisman Emergency Medicine

Institute at Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada; g St.
EP

h
Boniface General Hospital, University of Manitoba, Winnipeg, Manitoba, Canada; Halton
C

Healthcare Services, Oakville Hospital, Oakville, Ontario, Canada; i Robert and Dorothy Pitts
AC

Chair of Acute Care and Emergency Medicine, Keenan Research Centre, Li Ka Shing

Knowledge Institute, St Michael's Hospital, University of Toronto, Toronto, Ontario, Canada;

j k
Canadian Heart Research Centre, Toronto, Ontario, Canada; University Health Network and

Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada; l Population Health

Research Institute, Hamilton General Hospital, Hamilton, Ontario, Canada.

1
 ACCEPTED MANUSCRIPT

*The Trial of Routine Angioplasty and Stenting after Fibrinolysis to Enhance

Reperfusion in Acute Myocardial Infarction (TRANSFER-AMI) Investigators are listed

in the Appendix of Cantor et al N Engl J Med 2009;360:2705-18

PT
Running title: Long Term Outcomes of the Transfer-AMI study

RI
Word Count: 3,944

Conflicts of Interest: Yaron Arbel, Dennis Ko, Andrew Yan, Akshay Bagai, Maria Koh, Maria

SC
Eberg, and Mary Tan have no disclosures to report. Warren Cantor, David Fitchett, Bjug

U
Borgundvaag, John Ducas, Michael Heffernan, Laurie Morrison, Anatoly Langer, Vladimir
AN
Dzavik, Shamir Metha and Shaun Goodman have received research grant support from Canadian

Institutes of Health Research, Roche Canada, and Abbott Vascular Canada.

M

Corresponding author: Dr Shaun G. Goodman, St. Michael’s Hospital, Division of

D

Cardiology, 30 Bond Street, Room 6-034 Donnelly, Toronto, Ontario, Canada M5B 1W8; Fax
TE

416-864-5407 Telephone 416-864-5722. E-mail goodmans@smh.ca

C EP
AC

2
 ACCEPTED MANUSCRIPT

BRIEF SUMMARY

TRANSFER-AMI demonstrated superiority of routine early coronary angiography and

percutaneous coronary intervention at 30 days in patients presenting with STEMI and treated by

PT
fibrinolysis. In the present study, we evaluated the effects of the early invasive strategy after 7.8

RI
years using administrative datasets. There were no significant differences in death, myocardial

infarction (MI), unstable angina, stroke, transient ischemic attack or heart failure admissions

SC
between those randomized to an early invasive vs. standard treatment strategy.

U
AN
M
D
TE
C EP
AC

3
 ACCEPTED MANUSCRIPT

ABSTRACT

Background: The Trial of Routine Angioplasty and Stenting after Fibrinolysis to Enhance

Reperfusion in Acute Myocardial Infarction (TRANSFER-AMI) demonstrated superiority of

routine early coronary angiography (and percutaneous coronary intervention [PCI]) compared to

PT
standard therapy in fibrinolytic-treated ST-segment elevation myocardial infarction (STEMI)

RI
patients at 30 days. The aim of the present study was to evaluate the long term (>7 year) effects

of an early invasive strategy.

SC
Methods We linked the study cohort and administrative datasets to assess long term follow-up

U
status, including repeat procedures, hospitalizations, and major adverse cardiovascular events
AN
(MACE). Kaplan-Meier and Cox regression analysis were used to evaluate the relationship

between randomized treatment and long term adverse outcomes.

M
D

Results: A total of 881 patients had long term follow up and were included in our study. After a
TE

mean follow-up of 7.8 years, there were no significant differences in death, myocardial infarction

(MI), unstable angina, stroke, transient ischemic attack or heart failure admissions (HR 0.91
EP

[95% CI 0.73-1.13]; p=0.41) between those randomized to an early invasive vs. standard

treatment strategy. Following the index hospitalization, there were no significant difference in
C

the rates of coronary revascularization between the early invasive and the standard therapy
AC

groups (81 [19.3%] vs. 76 [17.9%]; p=0.61).

4
 ACCEPTED MANUSCRIPT

Conclusion: Despite the short-term benefit and safety of an early invasive strategy in STEMI

patients receiving fibrinolysis, no statistically-significant differences in MACE were observed

over 7.8 years.

PT
Keywords: Acute myocardial infarction, fibrinolysis, cardiac catheterization, percutaneous

RI
coronary intervention

U SC
AN
M
D
TE
C EP
AC

5
 ACCEPTED MANUSCRIPT

INTRODUCTION

Primary percutaneous coronary intervention (PCI) is the preferred method of reperfusion

for patients with ST-segment elevation myocardial infarction (STEMI).1, 2 However, many

patients with STEMI present to hospitals that do not have on-site capability and/or timely access

PT
to primary PCI. Thus, up to one third of patients presenting with STEMI continue to receive

RI
fibrinolysis as the initial mode of reperfusion.3-8 In the Trial of Routine Angioplasty and

Stenting after Fibrinolysis to Enhance Reperfusion in Acute Myocardial Infarction

SC
(TRANSFER-AMI),7 we randomly assigned high-risk STEMI patients who had received

fibrinolysis at Canadian hospitals that did not have the capability of performing PCI to either

U
standard treatment (including rescue PCI if required, or delayed angiography) or an early
AN
invasive strategy (immediate transfer to a PCI-capable hospital and coronary angiography and

PCI when appropriate within 6 hours post-fibrinolysis).5 At 30 days, there was a significant
M

reduction in the primary composite end point of death, reinfarction, recurrent ischemia, new or
D

worsening congestive heart failure, or cardiogenic shock in patients randomized to an early

TE

invasive strategy, with no increased bleeding compared to the standard therapy group. At one

year, rates of death or reinfarction, hospital readmission, and subsequent revascularization after
EP

the index hospitalization were similar between the invasive and standard arms.10 A meta-analysis

of randomized trials, including TRANSFER-AMI, suggested a lower rate of reinfarction at 6-12

C

months following the invasive strategy.14 The aim of the present study was to evaluate the
AC

longer term outcome of a standard versus early invasive strategy for a composite endpoint of

cardiovascular events including death, myocardial infarction and heart failure readmissions and

interventions.

6
 ACCEPTED MANUSCRIPT

METHODS

Study population

The design and primary results of the TRANSFER-AMI trial have been published.9, 10 In

brief, TRANSFER-AMI randomized 1059 patients (at 52 hospitals from Ontario, Manitoba, and

PT
Quebec from 2004-2007) receiving fibrinolysis for STEMI to an early invasive strategy,

RI
consisting of routine transfer for coronary angiography (and PCI where appropriate) within 6

hours of fibrinolysis, versus a standard approach (i.e., rescue angiography/PCI for failed lysis or

SC
elective angiography after 24 hours).

Patients receiving tenecteplase for STEMI at non-PCI capable hospitals who presented

U
within 12 hours of symptom onset were screened for eligibility. Participants met inclusion
AN
criteria if they had an anterior STEMI, or an inferior STEMI with at least one of the following

high-risk features: systolic blood pressure <100 mm Hg, heart rate >100 beats per minute, Killip
M

class II or III, ≥2 mm of ST depression in anterior leads, or right ventricular involvement (≥1

D

mm of ST elevation in right-sided lead V4). Key exclusion criteria included cardiogenic shock
TE

before randomization, PCI within the previous month, previous coronary-artery bypass surgery,

and availability of timely primary PCI (i.e., anticipated door-to-balloon time of less than 60
EP

minutes).

All participants received standard dose tenecteplase, aspirin, and unfractionated heparin
C

or Enoxaparin in the emergency department. Patients in the standard arm with persistent ST
AC

segment elevation and chest pain or with hemodynamic instability were transferred to a PCI-

capable center for rescue PCI; otherwise, patients in the standard arm remained at their

presenting hospital for at least 24 hours. The trial protocol recommended that all patients in the

7
 ACCEPTED MANUSCRIPT

standard arm undergo cardiac catheterization within two weeks of randomization, and was

amended in April 2005 to also strongly recommend the use of clopidogrel with fibrinolysis.

Ethics approval

The TRANSFER AMI trial required informed written consent (approved together with

PT
the protocol at each individual participating hospital) from all patients;11-13 this consent form, as

RI
per the original trial objectives, included provision of consent for follow-up to 1 year post

enrolment. Ethical approval for linkage of the Ontario TRANSFER-AMI patients (waiving the

SC
need for patient consent) was received from an independent ethics review board (Optimum

Clinical Research, Inc, Ethics Review Board, Ontario, Canada). This study was also approved by

U
the institutional review board at Sunnybrook Health Sciences Centre, Toronto, Canada.
AN
Data linkage

We attempted to link all Ontario patients randomized to the TRANSFER-AMI with

M

provincial administrative datasets to assess long term follow-up status, including repeat
D

procedures, hospitalizations, and major cardiovascular events (MACE). The Canadian Institute
TE

for Health Information (CIHI) Discharge Abstract Database was used to capture additional

comorbidities and subsequent hospitalizations. The Ontario Registered Persons Database was
EP

used to determine vital status of patients during follow-up. These datasets were linked using

unique encoded identifiers to protect patient confidentiality at the Institute for Clinical
C

Evaluative Sciences (ICES)

AC

Statistical Analysis

Descriptive statistics were used to describe the baseline characteristics, and categorical

variables were compared using a chi-squared test. Normally distributed continuous variables,

reported as mean and standard deviation, were compared using one-way ANOVA, while non-

8
 ACCEPTED MANUSCRIPT

normally distributed continuous variables, reported as median and interquartile range, were

compared using the Kruskal-Wallis test.

The primary efficacy endpoint of the long term analysis of the TRANSFER-AMI was the

composite of death, re-infarction, new unstable angina requiring hospitalization, new or

PT
worsening heart failure requiring hospitalization and TIA/stroke after at least 5 years of follow

RI
up since enrollment. Secondary endpoints included: coronary revascularization, admission for

stroke or TIA, and any hospitalization as well as a composite of death, re-infarction, new

SC
unstable angina requiring hospitalization, new or worsening heart failure requiring

hospitalization. We provided hazard ratios and 95% confidence intervals (CI) for all endpoints

U
analyzed. Kaplan-Meier curves were estimated for primary efficacy endpoint, all-cause
AN
mortality, and hospitalizations due to MI or CHF. In regression models, the Wald test was used

to assess the relationship between treatment and outcomes.

M

To control for potential baseline differences between the exposure groups, inverse
D

probability treatment weights (IPTW)3 were computed, with propensity scores derived by
TE

logistic modeling. We included the following variables in our propensity score model:

demographic characteristics (age, sex, indicator of residency in rural area, neighborhood income
EP

quintile, smoking status), previous comorbidities (prior congestive heart failure [CHF], prior

myocardial infarction [MI], prior PCI, prior stroke or transient ischemic attack [TIA], history of
C

hypertension, dyslipidemia, diabetes, cancer), and information measured prior to randomization

AC

(weight, systolic and diastolic blood pressure, heart rate, Killip class, location of ST-segment

elevation, time from symptom onset to tenecteplase administration and time from hospitalization

to tenecteplase administration). We used stabilized weights as described by P.Austin and

E.Stuart.14 For inverse-probability of treatment analyses, we present continuous variables using

9
 ACCEPTED MANUSCRIPT

weighted means and unweighted standard deviations, and categorical variables using weighted

percentages and unweighted frequencies. To quantitatively compare the balance achieved in

baseline covariates after weighting, we evaluated standardized differences after weighting with

IPTW: all baseline covariates had a standardized difference of less than 0.1. We used Kaplan-

PT
Meir curves and Cox regression analysis fitted to the weighted population to evaluate the

RI
relationship between exposure and long term adverse outcomes. Weighted log-rank test was used

to compare the survivor distributions of the two treatment groups.

SC
Statistical analyses were conducted using SAS 9.3 (SAS Institute, Cary, NC) program.

Two-sided p values <0.05 were considered statistically significant.

U
AN
RESULTS

Of 1,059 patients in the TRANSFER-AMI trial, including 952 patients (90%) in Ontario
M

that were randomized to an early invasive strategy or standard therapy, we were able to link and
D

determine long term follow up in 881 patients (92.5% of the original Ontario and 83.2% of the
TE

overall TRANSFER-AMI cohort) (Figure 1). We compared the clinical characteristics of

missing patients from Ontario, other provinces and the linked Ontario cohort, and found no
EP

significant differences (Table 1s). The mean age was 58.5 ± 11.7 years with 15.4% having

diabetes mellitus, 33% hypertension, and 27.2% dyslipidemia. Patient and procedural
C

characteristics are detailed in Tables 1 and 2.

AC

After a mean follow-up of 7.8 ±2.6 years (6,860 patient-years), there were no significant

differences between the early invasive and standard treatment strategies in the composite of in

death, myocardial infarction (MI), unstable angina, stroke, transient ischemic attack or heart

failure admissions (HR 0.91 [95% CI 0.73-1.13]; p=0.41) between those randomized to an early

10
 ACCEPTED MANUSCRIPT

invasive vs. standard treatment strategy. There was also no difference in death, MI, unstable

angina, or heart failure admission (HR 0.89 [95% CI 0.71-1.12]; p=0.32), or any of the

individual endpoints (Table 3 and Figure 2a-d). Similarly, there was no difference in the rates of

any hospital admission (HR 1.08 [0.9-1.28]; p= 0.41).

PT
Following the index hospitalization, there were no significant difference in the rates of

RI
coronary revascularization between the early invasive and the standard therapy groups (81

[19.3%] vs. 76 [17.9%]; p=0.61). (Table 4).

SC
In a sensitivity analysis, we compared the original cohorts with IPTW (n=872) and

observed similar results with no differences in the clinical characteristics (Table 1s), clinical

U
endpoints (Table 2s), or readmission rates (Table 3s).
AN
DISCUSSION
M

This is the first large scale randomized trial to report long term follow-up comparing an
D

early invasive strategy to standard therapy in STEMI patients receiving fibrinolysis. Despite
TE

short term benefits, after an average of almost 8 years of follow up, there were no significant

differences in major cardiovascular events between the two strategies.

EP

Early invasive studies

The early benefits of routine early PCI are to prevent re-occlusion of the infarct related
C

artery after successful reperfusion by fibrinolytic therapy and to improve myocardial function by
AC

reducing infarct size. In the early invasive strategy, the patient is transferred immediately for

routine early PCI rather than waiting to assess reperfusion success. At 30 days, the primary

endpoint of death, reinfarction, recurrent ischemia, heart failure or shock at 30 days was

significantly lower with the routine early PCI strategy.13 At one year, there was no statistically

11
 ACCEPTED MANUSCRIPT

significant difference in death or reinfarction between the 2 groups.11 Similarly, after almost 8

years, there were no significant differences in a primary composite outcome of death,

reinfarction, unstable angina, heart failure hospitalization, stroke, or TIA between the 2 treatment

strategies. In a meta-analysis that included TRANSFER-AMI, an early invasive strategy was

PT
associated with favorable outcomes up to 6-12 months.15 The benefits of routine invasive

RI
strategy over standard therapy were maintained at 6-12 months, with persistent reduction in the

endpoints of reinfarction (OR 0.64, 95% CI 0.40-0.98, p=0.01) and the composite of

SC
death/reinfarction (OR 0.71, 95% CI 0.52-0.97, p=0.03). 15 Clever et al. reported the results of

the long term follow up of the SIAM III Trial (n=197) 16 and after a mean follow up of 7.9 ± 3.4

U
years, an early (vs. a delayed) invasive strategy was associated with a significant reduction in the
AN
primary composite of death, reinfarction, target lesion revascularization or ischemic events (HR

0.61, 95% CI 0.42-0.88, p=0.008). Long-term survival was also numerically better in the early
M

invasive group (p=0.057). However, their cohort was small and only the composite endpoints
D

were found to be significantly different; thus, the play of chance cannot be excluded. In addition,
TE

the use of evidence-based, secondary prevention medications was lower in SIAM III (e.g., statins

60% vs. 89% in our study) and clopidogrel was not utilized; thus, the apparent benefit of the
EP

invasive strategy in SIAM III may have been magnified on a background of less evidence-based

therapy than used in the more contemporary TRANSFER AMI trial.

C

Despite the early benefit of routine early PCI in patients with STEMI, our analysis failed to
AC

identify longer term benefits of this strategy. The early benefit seen in the meta-analysis15 was

largely due to reduced reinfarction risk at one year. However, our follow up extends beyond one

year and recurrent ischemia and reinfarction often occurs in non-culprit coronary artery

territories.17 Thus, an early invasive strategy may have little to no impact on the risk of recurrent

12
 ACCEPTED MANUSCRIPT

ischemic events longer term. Patients achieving reperfusion with fibrinolysis may benefit from

reductions in recurrent ischemia or reinfarction with routine early PCI, while those who fail

reperfusion have a second opportunity for timely myocardial salvage. In addition, as other

studies have shown, some patients (high troponin, high GRACE scores) have elevated event

PT
rates regardless of treatment strategy.18, 19 The widespread use of ADP receptor inhibitor therapy

RI
(e.g., clopidogrel) that was used in our study may help reduce the thrombotic risks associated

with an early intervention after fibrinolysis.6, 20, 21 Therefore, medical therapy has a pivotal role

SC
in secondary prevention that transcends any differences seen in the first year.

Guidelines

U
According to the American guidelines, immediate transfer is indicated for cardiogenic
AN
shock or severe acute heart failure, irrespective of time delay from MI onset (Class I, LOE: B).

However, other clinical scenarios are less compelling (failed reperfusion or re-occlusion is
M

classified as Class IIa). In contrast, the ESC guidelines indicate that all patients should be
D

transferred (irrespective of baseline risk) after fibrinolysis (Class I, LOE: A). Our results do not
TE

detract from the previous findings of short-term benefits that informed guideline

recommendations; however, the longer term follow-up we now provide emphasizes that the
EP

benefits of an early invasive strategy in selected patients are likely limited to the short-term. 22

Given the numeric but not statistically-significant differences in MACE observed, we

C

estimated that 4,639 patients in each group would be required to demonstrate a significant
AC

difference (with 90% power) between the invasive and standard treatment groups; it is unlikely

that a study of this sample size will be undertaken.

There are several limitations to the current analysis. Data analysis was restricted to those patients

from the province of Ontario, and even with unique identifying patient information we were

13
 ACCEPTED MANUSCRIPT

unable to link the clinical trial and administrative databases for 8% of these patients. While we

did not find any baseline differences between patients from other provinces and those from

Ontario where database linkage was not successful and the analyzed Ontario cohort, it is possible

that the outcomes of these patients differed. The original study was powered for a primary

PT
composite endpoint that included events with a slightly different definition (e.g., new/worsening

RI
heart failure) or components (e.g., cardiogenic shock) not included in our long-term analysis.

Further, the current analysis is potentially underpowered, despite the prolonged follow up, to

SC
detect a potential benefit of an early invasive vs. standard strategy post-fibrinolysis. In addition,

there were some minor differences in the frequency of prescribed discharge medications (e.g.,

U
aspirin, beta blocker) between the groups; However, these differences were taken into
AN
consideration in the propensity analysis using inverse probability of treatment

weighting (IPTW).
M

Nevertheless, we did not find any trends for reduction in any of the endpoints evaluated.
D

Finally, this is a post hoc analysis. Despite these limitations, the present study represents the
TE

largest long term follow up report of an early invasive strategy after fibrinolysis compared with

standard treatment.
C EP

CONCLUSION
AC

Despite the short-term benefit and safety of an early invasive strategy in STEMI patients

receiving fibrinolysis, no differences in clinical outcome were observed over 7.8 years of follow

up.

14
 ACCEPTED MANUSCRIPT

Funding and Acknowledgements

The funding for the data linkage and analyses was provided by a Heart & Stroke Foundation of

Ontario/University of Toronto Polo Chair Award. The original TRANSFER-AMI trial was

supported by a grant from the Canadian Institutes of Health Research

PT
http://www.clinicaltrials.gov/ct2/show/NCT00164190; Unique Identifiers: NCT00164190 and an

RI
unrestricted research grant from Roche, Canada. Coronary stents were provided free of charge by

Abbott Vascular Canada. The funding sources had no involvement in the study conception or

SC
design; collection, analysis, and interpretation of data; in the writing, review, or approval of the

manuscript; and in the decision to submit the manuscript for publication.

U
This study was supported by the Institute for Clinical Evaluative Sciences (ICES), which
AN
is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care

(MOHLTC). The opinions, results and conclusions reported in this paper are those of the authors
M

and are independent from the funding sources. No endorsement by ICES or the Ontario
D

MOHLTC is intended or should be inferred. Parts of this material are based on data and
TE

information compiled and provided by CIHI. However, the analyses, conclusions, opinions and

statements expressed herein are those of the author, and not necessarily those of CIHI.
EP

Dr. Ko is supported by a Mid-Career Investigator Award from the Heart and Stroke

Foundation, Ontario office. Dr. Goodman is supported as the Heart and Stroke Foundation of
C

Ontario/University of Toronto (Polo) Chair.

AC

Disclosures

Yaron Arbel, Dennis Ko, Andrew Yan, Akshay Bagai, Maria Koh, Maria Eberg, and

Mary Tan have no disclosures to report. Warren Cantor, David Fitchett, Bjug Borgundvaag,

John Ducas, Michael Heffernan, Laurie Morrison, Anatoly Langer, Vladimir Dzavik, Shamir

15
 ACCEPTED MANUSCRIPT

Metha and Shaun Goodman have received research grant support from Canadian Institutes of

Health Research, Roche Canada, and Abbott Vascular Canada.

PT
RI
U SC
AN
M
D
TE
C EP
AC

16
 ACCEPTED MANUSCRIPT

References

1. Levine GN, Bates ER, Blankenship JC, et al. 2015 ACC/AHA/SCAI Focused Update on Primary

Percutaneous Coronary Intervention for Patients With ST-Elevation Myocardial Infarction: An

Update of the 2011 ACCF/AHA/SCAI Guideline for Percutaneous Coronary Intervention and the

PT
2013 ACCF/AHA Guideline for the Management of ST-Elevation Myocardial Infarction: A Report

RI
of the American College of Cardiology/American Heart Association Task Force on Clinical

Practice Guidelines and the Society for Cardiovascular Angiography and Interventions.

SC
Circulation. 2016;133:1135-1147.

2. Levine GN, Bates ER, Blankenship JC, et al. 2015 ACC/AHA/SCAI Focused Update on Primary

U
Percutaneous Coronary Intervention for Patients With ST-Elevation Myocardial Infarction: An
AN
Update of the 2011 ACCF/AHA/SCAI Guideline for Percutaneous Coronary Intervention and the

2013 ACCF/AHA Guideline for the Management of ST-Elevation Myocardial Infarction. Journal of
M

the American College of Cardiology. 2016;67:1235-1250.

D

3. Sim DS, Jeong MH, Ahn Y, et al. Pharmacoinvasive Strategy Versus Primary Percutaneous
TE

Coronary Intervention in Patients With ST-Segment-Elevation Myocardial Infarction: A

Propensity Score-Matched Analysis. Circulation. Cardiovascular interventions. 2016;9.

EP

4. Sigmundsson TS, Arnarson D, Rafnsson A, Magnusson V, Gunnarsson GT, Thorgeirsson G.

[Transport and treatment of patients with STEMI in rural Iceland--Only a few patients receive
C

PPCI within 120 minutes]. Laeknabladid. 2016;102:11-17.

AC

5. Chung SC, Sundstrom J, Gale CP, et al. Comparison of hospital variation in acute myocardial

infarction care and outcome between Sweden and United Kingdom: population based cohort

study using nationwide clinical registries. BMJ (Clinical research ed.). 2015;351:h3913.

17
 ACCEPTED MANUSCRIPT

6. Gandhi S, Zile B, Tan MK, et al. Increased uptake of guideline-recommended oral antiplatelet

therapy: insights from the Canadian acute coronary syndrome reflective. The Canadian journal

of cardiology. 2014;30:1725-1731.

7. Solla DJ, Paiva Filho Ide M, Delisle JE, et al. Integrated regional networks for ST-segment-

PT
elevation myocardial infarction care in developing countries: the experience of Salvador, Bahia,

RI
Brazil. Circulation. Cardiovascular quality and outcomes. 2013;6:9-17.

8. Lambert LJ, Brophy JM, Racine N, et al. Outcomes of Patients With ST-Elevation Myocardial

SC
Infarction Receiving and Not Receiving Reperfusion Therapy: The Importance of Examining All

Patients. The Canadian journal of cardiology. 2016;32:1325.e1311-1325.e1318.

9.

U
Cantor WJ, Fitchett D, Borgundvaag B, et al. Routine early angioplasty after fibrinolysis for acute
AN
myocardial infarction. N Engl J Med. 2009;360:2705-2718.

10. Cantor WJ, Fitchett D, Borgundvaag B, et al. Rationale and design of the Trial of Routine
M

ANgioplasty and Stenting After Fibrinolysis to Enhance Reperfusion in Acute Myocardial

D

Infarction (TRANSFER-AMI).[Erratum appears in Am Heart J. 2008 Feb;155(2):323 Note: Ducas,

John [added]]. American Heart Journal. 2008;155:19-25.

TE

11. Bagai A, Cantor WJ, Tan M, et al. Clinical outcomes and cost implications of routine early PCI
EP

after fibrinolysis: one-year follow-up of the Trial of Routine Angioplasty and Stenting after

Fibrinolysis to Enhance Reperfusion in Acute Myocardial Infarction (TRANSFER-AMI) study.

C

American heart journal. 2013;165:630-637.e632.

AC

12. Russo JJ, Goodman SG, Cantor WJ, et al. Efficacy and safety of a routine early invasive strategy in

relation to time from symptom onset to fibrinolysis (a subgroup analysis of TRANSFER-AMI). The

American journal of cardiology. 2015;115:1005-1012.

13. Cantor WJ, Fitchett D, Borgundvaag B, et al. Routine early angioplasty after fibrinolysis for acute

myocardial infarction. The New England journal of medicine. 2009;360:2705-2718.

18
 ACCEPTED MANUSCRIPT

14. Austin PC, Stuart EA. Moving towards best practice when using inverse probability of treatment

weighting (IPTW) using the propensity score to estimate causal treatment effects in

observational studies. Statistics in medicine. 2015;34:3661-3679.

15. Borgia F, Goodman SG, Halvorsen S, et al. Early routine percutaneous coronary intervention

PT
after fibrinolysis vs. standard therapy in ST-segment elevation myocardial infarction: a meta-

RI
analysis. European heart journal. 2010;31:2156-2169.

16. Clever YP, Cremers B, Link A, Bohm M, Scheller B. Long-term follow-up of early versus delayed

SC
invasive approach after fibrinolysis in acute myocardial infarction. Circulation. Cardiovascular

interventions. 2011;4:342-348.

17.

U
Stone GW, Maehara A, Lansky AJ, et al. A prospective natural-history study of coronary
AN
atherosclerosis. The New England journal of medicine. 2011;364:226-235.

18. Tan NS, Goodman SG, Cantor WJ, et al. Efficacy of Early Invasive Management After Fibrinolysis
M

for ST-Segment Elevation Myocardial Infarction in Relation to Initial Troponin Status. The
D

Canadian journal of cardiology. 2016;32:1221.e1211-1221.e1218.

Yan AT, Yan RT, Cantor WJ, et al. Relationship between risk stratification at admission and
TE

19.

treatment effects of early invasive management following fibrinolysis: insights from the Trial of
EP

Routine ANgioplasty and Stenting After Fibrinolysis to Enhance Reperfusion in Acute Myocardial

Infarction (TRANSFER-AMI). European heart journal. 2011;32:1994-2002.

C

20. Arbel Y, FitzGerald G, Yan AT, et al. Temporal trends in all-cause mortality according to smoking
AC

status: Insights from the Global Registry of Acute Coronary Events. Int J Cardiol. 2016;218:291-

297.

21. Arbel Y, Matetzky S, Gavrielov-Yusim N, et al. Temporal trends in all-cause mortality of smokers

versus non-smokers hospitalized with ST-segment elevation myocardial infarction. Int J Cardiol.

2014;176:171-176.

19
 ACCEPTED MANUSCRIPT

22. Bainey KR, Armstrong PW. Transatlantic Comparison of ST-Segment Elevation Myocardial

Infarction Guidelines: Insights From the United States and Europe. Journal of the American

College of Cardiology. 2016;67:216-229.

PT
RI
U SC
AN
M
D
TE
C EP
AC

20
 ACCEPTED MANUSCRIPT

Figures and Tables:

Table 1. Baseline characteristics of the study cohort

Table 2: Laboratory, Clinical and Procedural characteristics of the study cohort
Table 3. Outcomes in linked cohort (events from enrollment).

PT
Table 4: Revascularization (PCI or CABG) post-index hospitalization
Figure 1- Flow chart detailing the patient population
Figure 2- Kaplan-Meir curves for myocardial infarction (A), Death (B), CHF(C) and MACE (D)

RI
U SC
AN
M
D
TE
C EP
AC

21
 ACCEPTED MANUSCRIPT

Table 1. Baseline characteristics of the study cohort

Early Invasive Standard Overall Standardized

Characteristic N=444 N=437 N=881 Difference a P-value a
Follow-up, years

PT
Mean ± SD 7.6 ± 2.8 8.0 ± 2.5 7.8 ± 2.6 0.14 0.04
Age

RI
Mean ± SD 58.6 ± 11.6 58.4 ± 11.8 58.5 ± 11.7 0.02 0.77
>75 years, n (%) 42 (9.5%) 39 (8.9%) 81 (9.2%) 0.02 0.78

SC
Male, n (%) 352 (79.3%) 352 (80.5%) 704 (79.9%) 0.03 0.64
Prior congestive heart failure, n
(%) <=5 10 (2.3%) Suppressedb 0.16 0.02

U
Prior myocardial infarction, n (%) 47 (10.6%) 45 (10.3%) 92 (10.4%) 0.01 0.89

AN
Prior PCI, n (%) 28 (6.3%) 19 (4.3%) 47 (5.3%) 0.09 0.20
Prior stroke or transient ischemic
attack, n (%) 16 (3.6%) <=5 Suppressedb 0.16 0.02

M
History of smoking, n (%) 269 (61.1%) 262 (60.2%) 531 (60.7%) 0.02 0.78
Hypertension, n (%) 140 (31.5%) 151 (34.6%) 291 (33.0%) 0.06 0.34

D
Dyslipidemia, n (%) 116 (26.1%) 124 (28.4%) 240 (27.2%) 0.05 0.45
Diabetes, n (%) 65 (14.6%) 71 (16.2%) 136 (15.4%) 0.04 0.51

TE
Cancerc, n (%) 9 (2.0%) 11 (2.5%) 20 (2.3%) 0.03 0.63
Prior coronary catheterizationc, n
EP
(%) 22 (5.0%) 23 (5.3%) 45 (5.1%) 0.01 0.84
Length of stay in hospital
Mean ± SD 6.14 ± 7.06 7.09 ± 7.87 6.61 ± 7.48 0.13 0.057
C

Median (IQR) 4 (3-6) 5 (4-7) 5 (3-7) 0.4 <.001

AC

Killip class, n (%)

I 405 (91.6%) 401 (91.8%) 806 (91.7%) 0 0.224
II 28 (6.3%) 32 (7.3%) 60 (6.8%) 0.04
Abbreviations: CABG - coronary artery bypass graft surgery IQR – interquartile range; PCI - percutaneous coronary intervention; SD – standard deviation.
a
P-values and standardized differences are calculated based on the non-missing data.
b
Values are suppressed in accordance with ICES Re-identification Risk Assessment Procedure.
c
Information obtained from administrative data with lookback period of 5 years prior to randomization date.
 ACCEPTED MANUSCRIPT

Table 2: Laboratory, Clinical and Procedural characteristics of the study cohort

Early Invasive Standard Overall Standardized
Characteristic N=444 N=437 N=881 Difference a P-value a
Creatinine prior to randomization
(mg/dL)

PT
Median (IQR) 1 (1-1) 1 (1-1) 1 (1-1) 0.01 0.99
Elevated CK prior to randomization, n

RI
(%) 114 (30.5%) 100 (26.6%) 214 (28.5%) 0.09 0.214
Elevated Troponin (I/T) prior to
randomization, n (%) 164 (40.1%) 155 (38.1%) 319 (39.1%) 0.04 0.55

SC
GRACE score
Median (IQR) 122 (106-142) 126 (109-146) 123 (109-143) 0.1 0.15

U
Anterior Infarct, n (%) 251 (56.5%) 223 (51.0%) 474 (53.8%) 0.11 0.10

AN
With systolic BP <100 mm Hg, n (%) 45 (10.1%) 44 (10.1%) 89 (10.1%) 0 0.97
With heart rate >100 bpm, n (%) 24 (5.4%) 24 (5.5%) 48 (5.4%) 0 0.95
Time from symptom onset to

M
administration of tenecteplase (min)
Mean ± SD 162.6 ± 151.7 154.7 ± 130.7 158.7 ± 141.6 0.06 0.41

D
Median (IQR) 115 (74-185) 114 (75-190) 115 (75-188) 0.01 0.86
Time from hospital presentation to

TE
administration of tenecteplase (min) 39.9 ± 50.6 34.7 ± 36.9 37.4 ± 44.4 0.12 0.08
Mean ± SD
Median (IQR) 28 (17-44) 25 (16-41) 26 (16-42) 0.12 0.08
EP
Coronary catheterization, n (%) 436 (98.2%) 386 (88.3%) 822 (93.3%) 0.4 <.001
Percutaneous coronary intervention, n
C

(%) 377 (84.9%) 293 (67.0%) 670 (76.0%) 0.43 <.001

AC

Intra aortic balloon pump, n (%) 40 (9.0%) 21 (4.8%) 61 (6.9%) 0.17 0.01
Coronary-artery bypass grafting, n (%) 27 (6.1%) 33 (7.6%) 60 (6.8%) 0.06 0.39
d
Medications at discharge
Clopidogrel, n (%) 383 (91.2%) 347 (82.0%) 730 (86.6%) 0.27 <.001
Aspirin, n (%) 402 (95.7%) 390 (92.2%) 792 (94.0%) 0.15 0.03
Oral anticoagulant, n (%) 49 (11.7%) 36 (8.5%) 85 (10.1%) 0.1 0.13
Beta blocker, n (%) 380 (90.5%) 360 (85.1%) 740 (87.8%) 0.17 0.02
 ACCEPTED MANUSCRIPT

ACE inhibitor, n (%) 345 (82.1%) 335 (79.2%) 680 (80.7%) 0.08 0.28
Angiotensin receptor blocker, n (%) 7 (1.7%) 9 (2.1%) 16 (1.9%) 0.04 0.62
Statin, n (%) 377 (89.8%) 379 (89.6%) 756 (89.7%) 0.02 0.94
Diuretic, n (%) 30 (7.1%) 37 (8.7%) 67 (7.9%) 0.05 0.39

PT
Abbreviations: CABG - coronary artery bypass graft surgery IQR – interquartile range; PCI - percutaneous coronary intervention; SD –

RI
standard deviation.
a
P-values and standardized differences are calculated based on the non-missing data.

SC
b
Values are suppressed in accordance with ICES Re-identification Risk Assessment Procedure.
c
Information obtained from administrative data with lookback period of 5 years prior to randomization date.
d
Complete medication lists were available for 421 patients (invasive group) and 423 (standard group).

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 3. Outcomes in linked cohort (events from enrollment).

Event Exposure group Events, Person-Years Incidence Rate per Hazard Ratio P-
n (%) 100 PY (95% CI) valu
e

PT
Events at the end of follow-up
MACE (Death, MI, angina,

RI
CHF)
Standard (Reference) 159 (35.8%) 2,788.7 5.7 1
Early Invasive 152 (34.2%) 2,836.5 5.36 0.89 (0.71, 1.12) 0.32

SC
MACE(Death, MI, angina,
CHF and TIA/Stroke)

U
Standard (Reference) 167 (38.2%) 2,751.66 6.07 1

AN
Early Invasive 158 (35.6%) 2,851.06 5.54 0.91 (0.73, 1.13) 0.41
Death
Standard (Reference) 67 (15.3%) 3,481.66 1.92 1

M
Early Invasive 75 (16.9%) 3,378.82 2.22 1.15 (0.83, 1.60) 0.41
Myocardial re-infarction

D
Standard (Reference) 62 (14.2%) 3,101.89 2 1
Early Invasive 60 (13.5%) 3,114.13 1.93 0.95 (0.67, 1.36) 0.79

TE
Death or MI
Standard (Reference) 121 (27.7%) 3,101.89 3.9 1
EP
Early Invasive 120 (27.0%) 3,114.13 3.85 0.98 (0.76, 1.27) 0.89
TIA/Stroke
Standard (Reference) 16 (3.7%) 3,417.11 0.47 1
C

Early Invasive 17 (3.8%) 3,319.54 0.51 1.08 (0.55, 2.14) 0.69

AC

CHF
Standard (Reference) 46 (10.5%) 3,241.07 1.42 1
Early Invasive 35 (7.9%) 3,198.67 1.09 0.76 (0.49, 1.17) 0.21

Abbreviations: CHF - congestive heart failure; MI - myocardial infarction; PY - person-years; TIA - transient ischemic attack. MACE- Major
Cardiovascular events- a composite event of death/MI/ unstable angina/stroke/TIA/CHF
a
Values are suppressed in accordance with ICES Re-identification Risk Assessment Procedure.
 ACCEPTED MANUSCRIPT

Table 4: Revascularization (PCI or CABG) post-index hospitalization

Standard P-value
Early Invasive Treatment
Events, n (%) N=420 N=424

PT
30 days post discharge 9 (2.1%) 8 (1.9%) 0.79

RI
6 months post discharge 16 (3.8%) 22 (5.2%) 0.33
1 year post discharge 23 (5.5%) 27 (6.4%) 0.58

SC
3 years post discharge 37 (8.8%) 41 (9.7%) 0.66
5 years post discharge 55 (13.1%) 53 (12.5%) 0.79
Post discharge to end of follow-up 81 (19.3%) 76 (17.9%) 0.61

U
AN
Only events occurring after the index hospitalization were considered. Patients who died prior to discharge from the index
hospitalization were excluded.

M
** The denominator is restricted to those who survived the index hospitalization and did not have a PCI or CABG during the stay in
hospital.

D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC