Left Ventricular Ejection Fraction Reassessment Post-Myocardial Infarction:
Current Clinical Practice and Determinants of Adverse Remodeling

Derek S. Chew MD, Stephen B. Wilton MD, MSc, Katherine Kavanagh

MD, Danielle A. Southern MSc, Liong Eng Tan-Mesiatowsky MD, Derek V.
Exner MD, MPH

PII: S0002-8703(17)30381-2
DOI: doi: 10.1016/j.ahj.2017.11.014
Reference: YMHJ 5584

To appear in: American Heart Journal

Received date: 3 July 2017

Accepted date: 28 November 2017

Please cite this article as: Chew Derek S., Wilton Stephen B., Kavanagh Katherine,
Southern Danielle A., Tan-Mesiatowsky Liong Eng, Exner Derek V., Left Ventricu-
lar Ejection Fraction Reassessment Post-Myocardial Infarction: Current Clinical Prac-
tice and Determinants of Adverse Remodeling, American Heart Journal (2017), doi:
10.1016/j.ahj.2017.11.014

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

Left Ventricular Ejection Fraction Reassessment Post-Myocardial Infarction: Current

T
Clinical Practice and Determinants of Adverse Remodeling

P
RI
Derek S. Chew, MD; Stephen B. Wilton, MD, MSc; Katherine Kavanagh, MD; Danielle A.

SC
Southern, MSc; Liong Eng Tan-Mesiatowsky, MD; Derek V. Exner, MD, MPH; on behalf of the

APPROACH Investigators

NU
MA
Libin Cardiovascular Institute of Alberta

University of Calgary, Calgary, AB, Canada

ED

Corresponding Author:
PT

Dr. Derek Exner

GE63 TRW Building

CE

3280 Hospital Drive NW

Calgary, AB T2N 4Z6, CANADA

AC

Phone: (403) 220-3219 Fax: (403) 210-8140

Email: exner@ucalgary.ca

1
 ACCEPTED MANUSCRIPT

ABSTRACT

Background: Left ventricular (LV) dysfunction may be sustained or aggravated during the

convalescent months following an acute myocardial infarction (MI) and is difficult to predict. We

T
sought to determine current practice patterns of LV ejection fraction (LVEF) reassessment

P
during the months following MI, and evaluate the predictors and clinical significance of LVEF

RI
change in a prospective post-MI patient cohort.

SC
Methods: Patients with an acute MI between June 2010 to August 2014 were identified using

NU
the Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease registry.
MA
Patients with initial LV dysfunction (LVEF <40% with first MI or <45% with multiple MI events)

underwent a protocol-driven repeat LVEF assessment in follow-up if routine LVEF

reassessment was not performed.

ED
PT

Results: Of 5,964 MI patients, follow up LVEF assessments were attained for 442 of the 695

patients who had significant LV dysfunction. A sizable proportion (25%) had either no increase
CE

or a decline in LVEF. Adverse remodeling was associated with an anterior MI location, greater

peak serum troponin T, and a higher baseline LVEF at time of MI. Adverse LV remodeling
AC

conferred a 3-fold risk of death (HR 3.0, 95% CI 1.6 to 5.7, p=0.001), adjusted for baseline

LVEF, anterior MI location, and medication usage.

Conclusions: Current practice of LVEF reassessment during the convalescent months post-MI

is suboptimal, despite a sizeable proportion of patients that undergo adverse LV remodeling.

Targeting processes affecting low rates of LVEF reassessment may reduce missed care

opportunities and ensure that patients consistently receive appropriate evidence-based and

guideline-recommended care.

2
 ACCEPTED MANUSCRIPT

KEYWORDS

Myocardial Infarction

T
Left Ventricular Remodelling

P
LV Ejection Fraction

RI
SC
NU
MA
ED
PT
CE
AC

3
 ACCEPTED MANUSCRIPT

INTRODUCTION

Left ventricular (LV) dysfunction following an acute myocardial infarction (MI) identifies

patients at higher risk of sudden cardiac arrest and death,1-3 and remains an important predictor

T
of morbidity and mortality, even in an era of primary percutaneous coronary intervention.4, 5 The

P
change in LV function during the convalescent months following MI varies, and is difficult to

RI
predict. Improvement of LV function may occur early after MI due to recuperation of hibernating

SC
or stunned myocardium; however, the degree of long-term LV recovery is tempered by adverse

LV remodeling from myocyte necrosis, inflammation and fibrosis.5-8

NU
Up to 50% of patients do not demonstrate improvement in LV ejection fraction (EF)

during the months following acute MI despite revascularization and optimal medical therapy.9, 10
MA
The absence of LVEF recovery has prognostic significance; and is associated with an increased

risk of sudden cardiac arrest and all-cause mortality, independent of baseline LVEF.11
ED

Thus, it is important to determine which patients may have sustained or aggravated LV

PT

systolic dysfunction and the predictors of LVEF change. In addition to its prognostic

significance, predicting LVEF deterioration may have important therapeutic implications, such as
CE

qualifying patients for additional medical or device therapies. In the absence of reliable

predictors of adverse LV remodeling, current guidelines recommend LVEF reassessment during

AC

the convalescent months post-MI.12 Unfortunately, few studies have assessed the prevalence

of LVEF reassessment, but current practice is thought to variable and suboptimal. Patient-

reported LVEF reassessment at 6 months follow up was less than 40% in one cohort study that

included patients with significant LV dysfunction at time of MI (LVEF < 40%).13

The current study sought to evaluate the existing practice of LVEF assessment in routine

clinical practice, and identify the factors associated with LV remodeling in a contemporary

longitudinal, observational post-MI cohort from the Acute Myocardial Quality Assurance

(AMIQA) Study.

4
 ACCEPTED MANUSCRIPT

METHODS

Patient Population

The Acute Myocardial Infarction Quality Assurance (AMIQA) study (NCT#02399891) is a

T
prospective, observational cohort aimed at evaluating the incidence of LV dysfunction post-MI,

P
and assessing the prognostic utility of change in ejection fraction (EF) over the initial 12 months

RI
following MI. Patients were enrolled if they survived an acute MI (ST segment elevation

SC
myocardial infarction [STEMI] or non-ST elevation myocardial infarction [NSTEMI]) within

southern Alberta (approximate catchment population of 1.7 million people) and had evidence of

NU
a reduced LVEF at time of myocardial infarction (i.e. LVEF measured within two months of MI):
MA
defined as at least moderate LV dysfunction (LVEF < 0.40) with no prior history of MI, or mild-

moderate dysfunction (LVEF < 0.45) with prior MI events). This study was approved by the the

Conjoint Health Research Ethics Board of the University of Calgary, and conducted in
ED

accordance with the Declaration of Helsinki.

PT

Patients were identified through the Alberta Provincial Project for Outcome Assessment

in Coronary Heart Disease (APPROACH) database, which is a provincial registry with that
CE

prospectively collects detailed clinical information on all patients undergoing coronary

angiography in Alberta since 1995, and all patients admitted to a cardiac service in Southern
AC

Alberta since 2004. Data collection includes demographics, cardiac risk factors, comorbidities,

and procedural details of cardiac angiography and any subsequent revascularization

procedures.14 The accuracy of comorbidities and cardiac risk factors is verified through a data

enhancement procedure to ensure completeness of the database.15 Follow-up mortality data for

all patients in the database is attained through quarterly patient-level linkage to the Alberta

Bureau of Vital Statistics.

This project was conducted in 2 phases. First, to assess the rate of LVEF reassessment

in routine clinical practice, patients with index MI between June 2010 and November 2011 were

included. This included both passive (collection of clinical LVEF reassessment data) and active

5
 ACCEPTED MANUSCRIPT

(protocol-directed LVEF reassessment where clinical LVEF reassessment was not planned)

surveillance.

To determine factors associated with LVEF change, all patients meeting inclusion criteria

T
between December 2011 and August 2014 underwent protocol-directed LVEF reassessment.

P
Specifically, the patient’s family physician or primary cardiologist was reminded to arrange for

RI
reassessment of LVEF beyond during the convalescent months post-MI (i.e. within 2 to 12

SC
months after index myocardial infarction). The imaging modality used for EF reassessment was

left to the discretion of the ordering physician, although the majority of imaging was by

NU
echocardiography: 81% for initial LVEF assessment and 74% for follow up LVEF assessment
MA
(Supplemental Appendix 1). We prospectively collected all EF values during the initial 12

months of follow-up.

Funding
ED

No extramural funding was used to support this work.

PT

Statistical Analysis

Continuous variables are presented as median and interquartile ranges (IQR). Categorical data
CE

are expressed as frequencies and percentages. The Kruskal-Wallis test was used to compare

continuous variables, while the Pearson’s Chi-squared test was used to compare categorical
AC

variables. Multivariate logistic regression analysis was used to examine the predictors of LVEF

recovery.

Patients were stratified according to the degree of LVEF recovery by the following pre-

specified LVEF categories: decline or no recovery (LVEF Δ ≤ 0%), modest improvement (LVEF

Δ 1% to 9%), or large improvement (LVEF Δ > 10%). The ability of these EF recovery groups to

predict all-cause mortality was assessed using Cox multivariate models from which hazard

ratios and 95% confidence intervals (CI) were obtained. Kaplan-Meier time to event curves and

the differences in survival were assessed using the log-rank test statistic. Statistical tests were

2-sided with a p value less than 0.05 considered significant. Analyses were performed using

6
 ACCEPTED MANUSCRIPT

Stata/IC 13.1 (StataCorp, TX). The authors are solely responsible for the design and conduct of

this study, all study analyses, the drafting and editing of the paper and its final contents

T
RESULTS

P
LVEF Reassessment Rates in Routine Clinical Practice

RI
Between June 2010 and November 2011, 3,318 patients sustained a myocardial

SC
infarction. Of these patients, 321 (10%) had significant initial LV dysfunction using the study

definition and 152 (47%) underwent LVEF reassessment within 12 months per clinical practice.

NU
Of these 152 patients, 67 (44%) had persistent LV dysfunction (LVEF < 0.40) when reassessed
MA
(Figure 1). There were 169 patients in whom clinical LVEF reassessment was not planned and

a protocol-directed LVEF assessment undertaken; 23 patients (14%) were found to have an

LVEF < 0.40 remote to the index MI (p < 0.001 for clinical vs. protocol-identified LVEF < 0.40).
ED

Routine clinical practice was sub-optimal in identifying patients with sustained LV dysfunction
PT

(i.e. LVEF <0.40) with a sensitivity of 74% and specificity of 63%. All-cause mortality at 2 years

follow up was similar in patients undergoing either clinical or protocol-directed LVEF

CE

reassessment (5.9% vs. 5.3%, p = NS).

AC

Protocol-Directed LVEF Reassessment

Between December 2011 and August 2014, there were 5,964 MIs (Figure 2). There

were 695 patients (12%) with evidence of impaired LV systolic function. There were 253

patients that were excluded from the final analysis: 68 died prior to repeat LVEF assessment,

108 were either lost to follow up or had an out-of-province residence, 68 declined participation

(i.e. patient or primary physician refusal), and 9 were not included on basis of limited life-span

due to active medical comorbidities (e.g. palliative care patients). For data analysis, there were

442 patients with completed follow up LVEF measurements.

7
 ACCEPTED MANUSCRIPT

LVEF Recovery Post-MI

The overall median baseline ejection fraction was 35% (IQR 31% to 40%) assessed

within 1 day [IQR 0 to 3 days] post-MI. The median follow up LVEF was 43% (IQR 34 to 51%)

T
reassessed by 17 weeks (IQR 13 to 24 weeks) after MI. There were 128 patients (30%) who

P
had an LVEF ≤ 35% upon follow up assessment.

RI
Of the 442 patients, 109 (25%) had no recovery (LVEF Δ ≤ 0%), 150 (34%) had a

SC
modest improvement (LVEF Δ 1% to 10%) and 183 (41%) had large improvement (LVEF Δ

>10%). Interestingly, regardless of the initial baseline LVEF, there was a similar proportion of

NU
patients that underwent adverse remodeling, defined as a LVEF Δ ≤ 0% (Figure 3). An inverse
MA
relationship between baseline LVEF and LVEF recovery group was observed. Patients with no

LVEF recovery tended to have higher median baseline EFs (38%, IQR 35 to 40%) compared to

those with modest LVEF recovery (36%, IQR 33 to 40%) and large LVEF recovery (35%, IQR
ED

29 to 40%; p<0.001).
PT

Baseline Characteristics
CE

Clinical baseline characteristics, stratified by degree of EF recovery, are presented in

Table 1. The frequency of medical comorbidities among the three LVEF recovery groups were
AC

similar, except for prior history of myocardial infarction or coronary artery bypass graft surgery

(CABG).

There was no significant difference in frequency or modality of revascularization

between the EF recovery groups. Of the 108 patients that were treated with medical

management alone, PCI was not possible in 23 patients, medical therapy was thought to be the

best option in 22 patients, 17 patients were deemed too high risk for either CABG or PCI, 7

patients declined revascularization, 25 patients did not have viable myocardium in the territory

of occlusion, and the reasons were unclear for 12 patients.

8
 ACCEPTED MANUSCRIPT

Of the 411 patients surviving beyond 1 year post-MI, the rates of cardiac medication

usage (assessed at an average of 14.7 months post-index MI) were as follows: 85% aspirin,

36% P2Y12 inhibitor, 82% ACE-inhibitor or ARB, 81% beta-blocker, 21% aldosterone

T
antagonist, 19% oral anticoagulant, and 86% statin.

P
RI
Predictors of Adverse LV Remodeling

SC
The predictors of adverse LV remodeling (i.e. no LVEF recovery (Δ ≤ 0%) vs. modest or

large EF recovery) are listed in Table 2. By multivariate logistic regression analysis, an anterior

NU
location of myocardial infraction was an independent predictor for adverse LV remodeling with
MA
an odds ratio (OR) of 2.4 (95% CI, 1.3-4.5, p=0.007). Peak serum troponin T was also

associated with an increased risk of no LVEF recovery. An independent inverse relationship

was observed between baseline LVEF and adverse LV remodeling (OR 2.4 per 10% increase in
ED

LVEF; 95% CI 1.4–4.2; p=0.001) was observed. That is, a higher baseline LVEF at time of
PT

myocardial infarction was associated with an increased risk of no LVEF recovery.

CE

Adverse Remodeling and Clinical Outcomes

There were 43 deaths over a median follow up of 1.5 years (IQR 1.0 to 2.0 years).
AC

There was no significant difference in the median baseline LVEF between patients who died

(35%, IQR 32 to 40%) and those who did not (38%, IQR 29 to 40%; p=0.74). However,

mortality rates differed when stratified by degree of LVEF recovery. A linear trend in the all-

cause mortality was observed among patients with no recovery (18%), a moderate increase

(11%) and a large increase in LVEF (4%), respectively (p < 0.001) (Figure 4). Patients with no

LVEF recovery had a 3-fold higher risk of death (unadjusted HR 3.0, 95% CI 1.6 to 5.7,

p=0.001) compared to patients with a modest or large recovery in LVEF. Similar results were

observed after adjustment for important covariates (Table 3). Of note, the absence of EF

recovery was associated with an increase risk of all-cause mortality independent of baseline EF.

9
 ACCEPTED MANUSCRIPT

DISCUSSION

Main Findings

T
In a contemporary, prospective post-MI cohort of patients, we explored the role for

P
repeat LVEF assessment in the convalescent months post-MI. In particular, we determined the

RI
existing practice of LVEF reassessment post-MI, assessed the prognostic significance of LVEF

SC
recovery, and evaluated clinical, laboratory and angiographic predictors of LVEF change.

The main findings from this study were as follows: 1) in patients with significant LV

NU
systolic dysfunction post-MI, the rate of routine LVEF reassessment was less than 50% during
MA
the convalescent post-MI months; 2) absence of LVEF recovery was associated with a 3-fold

increased risk of all-cause mortality; and 3) predictors of adverse LV remodeling included an

anterior location of MI, greater peak serum troponin T, and a higher initial LVEF at time of MI.
ED
PT

Routine LVEF Reassessment Rates

A significant proportion of patients who suffer a MI will sustain or develop worsened LV

CE

systolic dysfunction in the months and years following their MI, which puts them at risk of a

number of adverse outcomes including heart failure, and sudden cardiac death. In our
AC

contemporary cohort of post-MI patients, 25% underwent adverse ventricular remodeling with

sustained or aggravated LV dysfunction. These rates are similar to earlier studies, despite the

improved availability of revascularization and evidence-based medical therapies.4, 10, 11, 16 It is

important to identify these high risk patients, as the detection of sustained or aggravated LV

systolic dysfunction may have important therapeutic implications. In our study, 30% of patients

had an LVEF ≤ 35% upon repeat LVEF assessment, who may be considered for additional

medical or device therapies.

The frequency of LVEF reassessment has not been well studied in the literature, but is

thought to be low despite recommendations from current guidelines. Reassessment of LV

10
 ACCEPTED MANUSCRIPT

function at least 40 days post-MI is a class I indication as per the current ACC/AHA STEMI

Guidelines especially in patients with reduced LVEF who are potential candidates for

implantable cardioverter defibrillator (ICD).12 The NSTEMI guidelines are less clear regarding

T
the need and frequency for LVEF reassessment post-MI.17 However, in patients who undergo

P
LVEF reassessment, there is an increased likelihood of appropriate ICD implantation.18

RI
In our study, only 47% of patients with initial LV dysfunction had an LVEF reassessment

SC
within 12 months as per routine clinical practice. These low rates are consistent with the

available literature. In one cohort study, the patient-reported LVEF reassessment rate was 35%

NU
within 6 months post-discharge among those patients with an LVEF < 40% at index

hospitalization.13 Lower rates of LVEF reassessment were associated with lack of medical
MA
insurance or involvement of a cardiologist during the index hospitalization for MI, despite similar

scheduled outpatient cardiology follow up.13 Interestingly, although 31% the 258 patients who
ED

were excluded from the analysis were lost to follow up, an additional 30% patients did not
PT

undergo repeat LVEF assessment due to their physician declining participation in the registry

during the protocol-driven phase of our study. Further study is required to explore the patient
CE

and provider factors contributing to sub-optimal LVEF reassessment; however, our results

suggest that the current clinical practice in determining the need for LVEF reassessment has
AC

poor sensitivity (74%) and specificity (63%) in predicting the patients with sustained LV systolic

dysfunction (LVEF < 40%).

Absence of LVEF Recovery Predicts Mortality

Baseline LVEF post-MI is a powerful predictor of cardiac morbidity and mortality.2, 19, 20 Few

studies have been performed to investigate the late recovery of LV function21, 22 and the existing

data supporting the use of LVEF change in the months post-MI to predict outcomes is even

more sparse.9, 10 Parodi et al. found that the 5-year cardiac mortality was over 2-fold higher

among post-MI patients without LVEF improvement (18%), compared to patients with a >10%

11
 ACCEPTED MANUSCRIPT

increased in LVEF over 6 months (8%; p=0.02).23 However, the lack of adjustment for important

covariates, and exclusion of high risk patients represent important limitations. In three

independent post-MI cohorts from Europe and Canada,24-26 the absence of LVEF recovery over

T
the initial 3 months was associated with a 6-fold increased risk of sudden cardiac arrest, and 4-

P
fold increase in all-cause mortality.27 These findings are congruent with the results of the

RI
current study, which found a 3-fold risk of death in post-MI patients that failed to demonstrate an

SC
improvement in LVEF. This risk was independent of revascularization, medications, and

baseline LVEF.

NU
The results of the current study add to the existing literature by validating the association

of LVEF recovery and mortality in a ‘real-world’ registry environment and in a more

MA
contemporary patient population.
ED

Predictors of LVEF Recovery

PT

The current study identified predictors of adverse LV remodeling, including peak serum

troponins, anterior location of infarction, and a higher baseline LVEF at time of myocardial
CE

infarction. Our data parallels work by Hallen et al. who found that a higher troponin I measured

within 24 to 48 hours post-MI was associated with attenuated LVEF improvement and increased
AC

LV volumes on echocardiography over a 4 month follow up period.28 However, similar to other

prior studies,29, 30 findings were limited by retrospective post-hoc analyses, narrow inclusion

criteria, and brief follow up. The usefulness of drawing inferences from those observations to

real-life contemporary post-MI management may be limited. In the present study, we describe a

prospective post-MI cohort of patients with peak serum troponin T as a predictor of sustained or

aggravated LV dysfunction. The prognostic value of peak serum cardiac biomarkers is

attributed to their ability to estimate infarct size,31-33 which has been studied as a powerful

predictor of adverse LV remodeling when quantified directly by scintigraphy or cardiac magnetic

resonance imaging.34, 35

12
 ACCEPTED MANUSCRIPT

Consistent with prior studies,36, 37 anterior location of MI was also found to be

independently predictive of adverse LV remodeling. An anterior location of MI has been shown

to have greater irreversible ischemic LV damage, mainly due to the greater magnitude of

T
myocardium at risk intrinsic to anterior infarcts.38 However, there has been some dispute

P
whether the adverse post-MI LV remodeling is primarily due to a larger infarct size, or whether

RI
there is an independent contribution to remodeling and prognosis based on infarct location.38-40

SC
In our study, both anterior location and peak serum troponin were independent predictors of

adverse LV remodeling. This suggests an independent contribution of anterior location in

NU
addition to infarct size, as estimated by peak serum troponin levels. Unfortunately, only limited
MA
conclusions can be drawn, as we did not collect systematic information on infarct size.

Interestingly, there was an inverse relationship between baseline LVEF and adverse LV

remodeling; that is, patients with a lower baseline LVEF at time of MI showed the greatest
ED

improvement in LV function. Reasons for this inverse relationship between baseline LVEF and
PT

LV function improvement are not entirely clear. However, it is possible that patients with a lower

initial LVEF have greater amounts of stunned myocardium, and subsequently a greater potential
CE

for functional recuperation.11 Christian et al. reported that 25% of their post-MI patients had a

lower than expected LVEF expected by scintigraphic estimates of their infarct size due to
AC

myocardial stunning.41 These patients had a late improvement in LVEF at 6 weeks post-MI.

Alternatively, our observation could simply represent regression toward the mean, as has been

reported previously.42

Limitations

The study results must be interpreted in the context of some inherent limitations. There were

several methods used for the LVEF assessment, which may cause some variation in the

accuracy of LVEF measurements. The majority of LVEF measurements were attained through

echocardiography, which has inherent technical limitations based on the adequacy of

13
 ACCEPTED MANUSCRIPT

endocardium visualization. An additional limitation was the failure to capture a precise LVEF in

all patients due to physician preference in reporting LVEF numerically or categorically. These

limitations must be interpreted in the context of attaining real-world registry based data, and the

T
pragmatism of acquiring LVEF assessments that reflect clinical practice. Unfortunately, a

P
significant number of patients were not included in the analysis due to failure to attain a repeat

RI
LVEF assessment. Finally, the EF recovery was treated categorically instead of as a

SC
continuum. However, the categories chosen allow for simplicity in clinical use, and as shown by

the present study, carry clinical relevance.

NU
CONCLUSIONS
MA
In this contemporary cohort study, less than half of patients with significant LV dysfunction post

MI had routine LVEF reassessment within 12 months. Using a prospective system of physician
ED

reminders and facilitation of testing, the proportion of these patients undergoing post-MI LVEF
PT

reassessment was increased. When reassessed, one quarter of patients had sustained or

exacerbated LVEF reduction despite contemporary management, a finding associated with a 3-

CE

fold increase in total mortality. Together, these findings support the importance of LVEF

reassessment beyond 3 months post-MI in those with acute LV dysfunction. Targeting

AC

processes affecting low rates of LVEF reassessment may reduce missed care opportunities and

ensure that patients consistently receive appropriate evidence-based and guideline-

recommended care.

14
 ACCEPTED MANUSCRIPT

References

1. Hammermeister KE, DeRouen TA and Dodge HT. Variables predictive of survival in

patients with coronary disease. Selection by univariate and multivariate analyses from the

T
clinical, electrocardiographic, exercise, arteriographic, and quantitative angiographic

P
evaluations. Circulation. 1979;59:421-430.

RI
2. The Multicenter Postinfarction Research G. Risk stratification and survival after

SC
myocardial infarction. The New England journal of medicine. 1983;309:331-336.

3. White HD, Norris RM, Brown MA, Brandt PW, Whitlock RM and Wild CJ. Left ventricular

NU
end-systolic volume as the major determinant of survival after recovery from myocardial

infarction. Circulation. 1987;76:44-51.

MA
4. Halkin A, Stone GW, Dixon SR, Grines CL, Tcheng JE, Cox DA, Garcia E, Brodie B,

Stuckey TD, Mehran R and Lansky AJ. Impact and determinants of left ventricular function in
ED

patients undergoing primary percutaneous coronary intervention in acute myocardial infarction.

PT

The American Journal of Cardiology. 2005;96:325-331.

5. Solomon SD, Skali H, Anavekar NS, Bourgoun M, Barvik S, Ghali JK, Warnica JW,
CE

Khrakovskaya M, Arnold JM, Schwartz Y, Velazquez EJ, Califf RM, McMurray JV and Pfeffer

MA. Changes in ventricular size and function in patients treated with valsartan, captopril, or both
AC

after myocardial infarction. Circulation. 2005;111:3411-3419.

6. Wijns W, Vatner SF and Camici PG. Hibernating myocardium. N Engl J Med.

1998;339:173-81.

7. Camici PG, Prasad SK and Rimoldi OE. Stunning, hibernation, and assessment of

myocardial viability. Circulation. 2008;117:103-114.

8. Braunwald E and Kloner RA. The stunned myocardium: prolonged, postischemic

ventricular dysfunction. Circulation. 1982;66:1146-1149.

9. Solomon SD, Glynn RJ, Greaves S, Ajani U, Rouleau JL, Menapace F, Arnold JM,

Hennekens C and Pfeffer MA. Recovery of ventricular function after myocardial infarction in the

15
 ACCEPTED MANUSCRIPT

reperfusion era: the healing and early afterload reducing therapy study. Annals of Internal

Medicine. 2001;134:451-458.

10. Ottervanger JP, van 't Hof AW, Reiffers S, Hoorntje JC, Suryapranata H, de Boer MJ

T
and Zijlstra F. Long-term recovery of left ventricular function after primary angioplasty for acute

P
myocardial infarction. European heart journal. 2001;22:785-790.

RI
11. Ndrepepa G, Mehilli J, Martinoff S, Schwaiger M, Schomig A and Kastrati A. Evolution of

SC
left ventricular ejection fraction and its relationship to infarct size after acute myocardial

infarction. Journal of the American College of Cardiology. 2007;50:149-156.

NU
12. O'Gara PT, Kushner FG, Ascheim DD, Casey DE, Jr., Chung MK, de Lemos JA, Ettinger
MA
SM, Fang JC, Fesmire FM, Franklin BA, Granger CB, Krumholz HM, Linderbaum JA, Morrow

DA, Newby LK, Ornato JP, Ou N, Radford MJ, Tamis-Holland JE, Tommaso CL, Tracy CM,

Woo YJ, Zhao DX, Anderson JL, Jacobs AK, Halperin JL, Albert NM, Brindis RG, Creager MA,
ED

DeMets D, Guyton RA, Hochman JS, Kovacs RJ, Kushner FG, Ohman EM, Stevenson WG,
PT

Yancy CW and American College of Cardiology Foundation/American Heart Association Task

Force on Practice G. 2013 ACCF/AHA guideline for the management of ST-elevation

CE

myocardial infarction: a report of the American College of Cardiology Foundation/American

Heart Association Task Force on Practice Guidelines. Circulation. 2013;127:e362-425.

AC

13. Miller AL, Gosch K, Daugherty SL, Rathore S, Peterson PN, Peterson ED, Ho PM, Chan

PS, Lanfear DE, Spertus JA and Wang TY. Failure to reassess ejection fraction after acute

myocardial infarction in potential implantable cardioverter/defibrillator candidates: insights from

the Translational Research Investigating Underlying disparities in acute Myocardial infarction

Patients' Health Status (TRIUMPH) registry. Am Heart J. 2013;166:737-43.

14. Ghali WA and Knudtson ML. Overview of the Alberta Provincial Project for Outcome

Assessment in Coronary Heart Disease. On behalf of the APPROACH investigators. Can J

Cardiol. 2000;16:1225-30.

16
 ACCEPTED MANUSCRIPT

15. Faris PD, Ghali WA, Brant R, Norris CM, Galbraith PD, Knudtson ML and Disease

AIAPPfOAiCH. Multiple imputation versus data enhancement for dealing with missing data in

observational health care outcome analyses. Journal of clinical epidemiology. 2002;55:184-91.

T
16. Bauters C, Fertin M, Delhaye C, Goeminne C, Le Tourneau T, Lamblin N and de Groote

P
P. Late recovery in left ventricular systolic function after discharge of patients with a first anterior

RI
myocardial infarction. Arch Cardiovasc Dis. 2010;103:538-45.

SC
17. Amsterdam EA, Wenger NK, Brindis RG, Casey DE, Jr., Ganiats TG, Holmes DR, Jr.,

Jaffe AS, Jneid H, Kelly RF, Kontos MC, Levine GN, Liebson PR, Mukherjee D, Peterson ED,

NU
Sabatine MS, Smalling RW, Zieman SJ, American College of C, American Heart Association
MA
Task Force on Practice G, Society for Cardiovascular A, Interventions, Society of Thoracic S

and American Association for Clinical C. 2014 AHA/ACC Guideline for the Management of

Patients with Non-ST-Elevation Acute Coronary Syndromes: a report of the American College of
ED

Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol.

PT

2014;64:e139-228.

18. Pokorney SD, Miller AL, Chen AY, Thomas L, Fonarow GC, de Lemos JA, Al-Khatib SM,
CE

Velazquez EJ, Peterson ED and Wang TY. Reassessment of Cardiac Function and Implantable

Cardioverter-Defibrillator Use Among Medicare Patients With Low Ejection Fraction After
AC

Myocardial Infarction. Circulation. 2017;135:38-47.

19. Dagres N and Hindricks G. Risk stratification after myocardial infarction: is left ventricular

ejection fraction enough to prevent sudden cardiac death? Eur Heart J. 2013;34:1964-71.

20. Rouleau JL, Talajic M, Sussex B, Potvin L, Warnica W, Davies RF, Gardner M, Stewart

D, Plante S, Dupuis R, Lauzon C, Ferguson J, Mikes E, Balnozan V and Savard P. Myocardial

infarction patients in the 1990s--their risk factors, stratification and survival in Canada: the

Canadian Assessment of Myocardial Infarction (CAMI) Study. J Am Coll Cardiol. 1996;27:1119-

27.

17
 ACCEPTED MANUSCRIPT

21. Antoni ML, Mollema SA, Atary JZ, Borleffs CJ, Boersma E, van de Veire NR, Holman

ER, van der Wall EE, Schalij MJ and Bax JJ. Time course of global left ventricular strain after

acute myocardial infarction. European heart journal. 2010;31:2006-2013.

T
22. Funaro S, La Torre G, Madonna M, Galiuto L, Scara A, Labbadia A, Canali E, Mattatelli

P
A, Fedele F, Alessandrini F, Crea F, Agati L and Investigators A. Incidence, determinants, and

RI
prognostic value of reverse left ventricular remodelling after primary percutaneous coronary

SC
intervention: results of the Acute Myocardial Infarction Contrast Imaging (AMICI) multicenter

study. European heart journal. 2009;30:566-575.

NU
23. Parodi G, Memisha G, Carrabba N, Signorini U, Migliorini A, Cerisano G and Antoniucci
MA
D. Prevalence, predictors, time course, and long-term clinical implications of left ventricular

functional recovery after mechanical reperfusion for acute myocardial infarction. The American

Journal of Cardiology. 2007;100:1718-1722.

ED

24. Exner DV, Kavanagh KM, Slawnych MP, Mitchell LB, Ramadan D, Aggarwal SG,
PT

Noullett C, Van Schaik A, Mitchell RT, Shibata MA, Gulamhussein S, McMeekin J, Tymchak W,

Schnell G, Gillis AM, Sheldon RS, Fick GH, Duff HJ and Investigators R. Noninvasive risk
CE

assessment early after a myocardial infarction the REFINE study. Journal of the American

College of Cardiology. 2007;50:2275-2284.

AC

25. Bauer A, Barthel P, Schneider R, Ulm K, Muller A, Joeinig A, Stich R, Kiviniemi A,

Hnatkova K, Huikuri H, Schomig A, Malik M and Schmidt G. Improved Stratification of

Autonomic Regulation for risk prediction in post-infarction patients with preserved left ventricular

function (ISAR-Risk). European heart journal. 2009;30:576-583.

26. Huikuri HV, Raatikainen MJ, Moerch-Joergensen R, Hartikainen J, Virtanen V, Boland J,

Anttonen O, Hoest N, Boersma LV, Platou ES, Messier MD, Bloch-Thomsen PE, Cardiac A and

Risk Stratification after Acute Myocardial Infarction study g. Prediction of fatal or near-fatal

cardiac arrhythmia events in patients with depressed left ventricular function after an acute

myocardial infarction. European heart journal. 2009;30:689-698.

18
 ACCEPTED MANUSCRIPT

27. Chew D, Huikuri H, Schmidt G, Kavanagh K, Dommasch M, Thomsen PEB, Sinnecker

D, Raatikainen P and Exner D. The Degree of Left Ventricular Ejection Fraction Change

Following Myocardial Infarction Predicts Risk of Sudden Cardiac Death (abstract). Journal of the

T
American College of Cardiology. 2015;65.

P
28. Hallen J, Jensen JK, Fagerland MW, Jaffe AS and Atar D. Cardiac troponin I for the

RI
prediction of functional recovery and left ventricular remodelling following primary percutaneous

SC
coronary intervention for ST-elevation myocardial infarction. Heart (British Cardiac Society).

2010;96:1892-1897.

NU
29. Fertin M, Hennache B, Hamon M, Ennezat PV, Biausque F, Elkohen M, Nugue O, Tricot
MA
O, Lamblin N, Pinet F and Bauters C. Usefulness of serial assessment of B-type natriuretic

peptide, troponin I, and C-reactive protein to predict left ventricular remodeling after acute

myocardial infarction (from the REVE-2 study). Am J Cardiol. 2010;106:1410-6.

ED

30. Hallen J. Troponin for the estimation of infarct size: what have we learned? Cardiology.
PT

2012;121:204-212.

31. Arruda-Olson AM, Roger VL, Jaffe AS, Hodge DO, Gibbons RJ and Miller TD. Troponin
CE

T levels and infarct size by SPECT myocardial perfusion imaging. JACC Cardiovasc Imaging.

2011;4:523-33.
AC

32. Younger JF, Plein S, Barth J, Ridgway JP, Ball SG and Greenwood JP. Troponin-I

concentration 72 h after myocardial infarction correlates with infarct size and presence of

microvascular obstruction. Heart. 2007;93:1547-51.

33. Mollema SA, Liem SS, Suffoletto MS, Bleeker GB, van der Hoeven BL, van de Veire NR,

Boersma E, Holman ER, van der Wall EE, Schalij MJ, Gorcsan J, 3rd and Bax JJ. Left

ventricular dyssynchrony acutely after myocardial infarction predicts left ventricular remodeling.

J Am Coll Cardiol. 2007;50:1532-40.

34. Pfeffer MA and Braunwald E. Ventricular remodeling after myocardial infarction.

Experimental observations and clinical implications. Circulation. 1990;81:1161-1172.

19
 ACCEPTED MANUSCRIPT

35. Nijveldt R, Beek AM, Hirsch A, Stoel MG, Hofman MB, Umans VA, Algra PR, Twisk JW

and van Rossum AC. Functional recovery after acute myocardial infarction: comparison

between angiography, electrocardiography, and cardiovascular magnetic resonance measures

T
of microvascular injury. J Am Coll Cardiol. 2008;52:181-9.

P
36. Hands ME, Lloyd BL, Robinson JS, de Klerk N and Thompson PL. Prognostic

RI
significance of electrocardiographic site of infarction after correction for enzymatic size of

SC
infarction. Circulation. 1986;73:885-91.

37. Stone PH, Raabe DS, Jaffe AS, Gustafson N, Muller JE, Turi ZG, Rutherford JD, Poole

NU
WK, Passamani E, Willerson JT and et al. Prognostic significance of location and type of
MA
myocardial infarction: independent adverse outcome associated with anterior location. J Am Coll

Cardiol. 1988;11:453-63.

38. Masci PG, Ganame J, Francone M, Desmet W, Lorenzoni V, Iacucci I, Barison A,

ED

Carbone I, Lombardi M, Agati L, Janssens S and Bogaert J. Relationship between location and
PT

size of myocardial infarction and their reciprocal influences on post-infarction left ventricular

remodelling. Eur Heart J. 2011;32:1640-8.

CE

39. Erlebacher JA, Weiss JL, Weisfeldt ML and Bulkley BH. Early dilation of the infarcted

segment in acute transmural myocardial infarction: role of infarct expansion in acute left
AC

ventricular enlargement. J Am Coll Cardiol. 1984;4:201-8.

40. Thanavaro S, Kleiger RE, Province MA, Hubert JW, Miller JP, Krone RJ and Oliver GC.

Effect of infarct location on the in-hospital prognosis of patients with first transmural myocardial

infarction. Circulation. 1982;66:742-7.

41. Christian TF, Behrenbeck T, Pellikka PA, Huber KC, Chesebro JH and Gibbons RJ.

Mismatch of left ventricular function and infarct size demonstrated by technetium-99m isonitrile

imaging after reperfusion therapy for acute myocardial infarction: identification of myocardial

stunning and hyperkinesia. Journal of the American College of Cardiology. 1990;16:1632-1638.

20
 ACCEPTED MANUSCRIPT

42. Berger PB and Gersh BJ. Ventricular function after primary angioplasty for acute

myocardial infarction: correlates and caveats. Eur Heart J. 2001;22:726-8.

P T
RI
SC
NU
MA
ED
PT
CE
AC

21
 ACCEPTED MANUSCRIPT

Figures and Figure Legend

Figure 1. Current Practice of LVEF Reassessment for patients with index MI occurring between
June 2010 to November 2011

P T
RI
SC
NU
MA
ED
PT
CE
AC

22
 ACCEPTED MANUSCRIPT

Figure 2. Prospective project-directed LVEF reassessment for patients with index MI occurring
between December 2011 to August 2014

P T
RI
SC
NU
MA
ED
PT
CE
AC

23
 ACCEPTED MANUSCRIPT

Figure 3. Change in Left Ventricular Ejection Fraction Stratified by Degree of Initial LV Systolic
Dysfunction.

P T
RI
SC
NU
MA
ED
PT
CE
AC

24
 ACCEPTED MANUSCRIPT

Figure 4. Risk of All Cause Death Stratified by LVEF Recovery Groups

P T
RI
SC
NU
MA
ED
PT
CE
AC

25
 ACCEPTED MANUSCRIPT

Table 1. Baseline Characteristics, Stratified by EF Recovery Group

LVEF Change

EF< 0% 0<EF<10% EF >10% P value

T
(n = 109) (n = 150) (n = 183)

P
Age (years) 67 [59, 75] 64 [56, 73] 63 [53, 74] 0.04

RI
Female (%) 19 15 28 0.01

SC
BMI (kg/m2) 28 [25, 30] 28 [25, 32] 27 [24, 31] 0.2

Median Baseline EF (%) 38 [35, 40] 36 [33, 40] 35 [29, 40] 0.0001

NU
Median Change in EF (%) -5 [-10, -2]
MA +6 [+3, +8] +18 [+14, 0. 0001

+23]
ED

Vascular Risk Factors

Hypertension (%) 62 57 50 0.2

PT

Smoking – Any (%) 50 59 54 0.4

Current Smoker 29 33 32 0.8

CE

Prior Smoker 21 26 21 0.5

AC

Diabetes (%) 33 34 27 0.4

Kidney Disease (%) 9 9 4 0.1

Prior Cardiac History

Prior CABG (%) 15 9 3 0.002

Prior PCI (%) 35 33 24 0.07

Prior MI (%) 43 31 28 0.02

History of CHF (%) 14 14 8 0.2

26
 ACCEPTED MANUSCRIPT

Management

PCI (%) 60 61 70 0.1

CABG (%) 10 14 9 0.3

T
Thrombolysis (%) 15 13 9 0.3

P
Medical Management (%) 30 25 21 0.2

RI
SC
Index MI Details

Anterior Location (%) 70 55 48 0.001

NU
STEMI (%) 63 60 55
0.4
MA
NSTEMI (%) 37 40 45

CAD Extent (0-1/2/3) (%) 19 / 18 / 63 23 / 26 / 51 32 / 27 / 41 0.005

ED

LVEDP (mmHg) 27 [19, 32] 24, [18, 30] 24 [17, 30] 0.2

Cardiac Arrest at time of MI 13 7 7 0.1

PT

(%)
CE

Laboratory Data
AC

Baseline Cr (umol/L) 89 [75, 105] 89 [75, 109] 83 [70, 97] 0.01

Baseline LDL (mmol/L) 2.2 [1.6, 2.8] 2.1 [1.5, 3.0] 2.3 [1.7, 3.0] 0.08

Peak Troponin T (mcg/L) 2.7 [0.7, 10.2] 3.0 [0.6, 8.1] 1.9 [0.6, 5.6] 0.04

Discharge Medications

ASA (%) 99 99 99 0.8

P2Y12 inhibitor (%) 80 89 86 0.09

ACE inhibitor / ARB (%) 94 93 93 0.9

Beta-Blocker (%) 92 94 95 0.5

27
 ACCEPTED MANUSCRIPT

Statin (%) 95 95 97 0.7

Aldosterone Antagonist (%) 21 19 21 0.9

Warfarin (%) 48 49 39 0.2

TP
RI
SC
NU
MA
ED
PT
CE
AC

28
 ACCEPTED MANUSCRIPT

Table 2. Predictors of Adverse Remodeling in uni- and multivariate logistic regression

Univariate Multivariate

OR 95% CI P OR 95% CI P

T
Age (per decade) 1.3 1.1 – 1.5 0.01 -- -- --

P
Female -- -- NS -- -- --

RI
Diabetes -- -- NS -- -- --

SC
History of Prior MI 1.8 1.2 – 2.9 0.007 -- -- --

History of Prior CABG 2.7 1.4 – 5.5 0.005 -- -- --

NU
Anterior MI Location 2.2 1.4 – 3.5 0.001 2.4 1.3 – 4.5 0.007
MA
CAD Extent 1.5 1.1 – 2.0 0.005 -- -- --

Peak Troponin T (per 5.0 1.1 1.04 – 1.20 0.01 1.4 1.1 – 1.7 0.006
ED

mcg/L)

Baseline Creatinine -- -- NS -- -- --
PT

Baseline EF (per 10%) 1.9 1.3 – 2.8 0.001 2.4 1.4 – 4.2 0.001
CE
AC

29
 ACCEPTED MANUSCRIPT

Table 3. “No EF Recovery” as Predictor of All Cause Death

OR 95% CI P

Model 1 (No EF Recovery) 3.0 1.6 – 5.7 0.001

T
Model 2 (No EF Recovery, Anterior 3.0 1.5 – 5.9 0.001

P
Location, baseline EF)

RI
Model 3 (No EF Recovery, Anterior 3.0 1.5 – 5.8 0.002

SC
Location, baseline EF, and Lack of

Beta-Blocker or ACE-I/ARB)

NU
MA
ED
PT
CE
AC

30