Author + information
- Received April 28, 2017
- Revision received December 22, 2017
- Accepted December 28, 2017
- Published online February 28, 2018.
- Derek S. Chew, MDa,
- Huikuri Heikki, MDb,
- Georg Schmidt, MDc,
- Katherine M. Kavanagh, MDa,
- Michael Dommasch, MDc,
- Poul Erik Bloch Thomsen, MD, DMSCb,
- Daniel Sinnecker, MDc,
- Pekka Raatikainen, MDa and
- Derek V. Exner, MD, MPHa,∗ ()
- aLibin Cardiovascular Institute of Alberta, University of Calgary, Calgary, Alberta, Canada
- bDepartment of Internal Medicine, Division of Cardiology, University of Oulu, Oulu, Finland
- cMedizinische Klinik und Poliklinik, Klinikum rechts der Isar, Technische Universität München, Munich, Germany
- ↵∗Address for correspondence:
Dr. Derek Exner, Libin Cardiovascular Institute of Alberta, University of Calgary, GE63 TRW Building, 3280 Hospital Drive NW, Calgary, AB T2N 4Z6, Canada.
Objectives This study hypothesizes that a lack of left ventricular ejection fraction (LVEF) recovery after myocardial infarction (MI) would be associated with a poor outcome.
Background A reduced LVEF early after MI identifies patients at risk of adverse outcomes. Whether the change in LVEF in the weeks to months following MI provides additional information on prognosis is less certain.
Methods Change in LVEF between the early (2 to 7 days) and later (2 to 12 weeks) post-MI periods in patients with a first MI was assessed in 3 independent cohorts (REFINE [Risk Estimation Following Infarction Noninvasive Evaluation]; CARISMA [Cardiac Arrhythmia and Risk Stratification after Myocardial Infarction]; ISAR [Improved Stratification of Autonomy Regulation]). Patients were categorized as having no recovery (Δ ≤ 0%), a modest increase (Δ 1% to 9%), or a large increase (Δ ≥ 10%) in LVEF. The relationship between change in LVEF and risk of sudden cardiac arrest (SCA) and all-cause mortality were assessed in Cox multivariable models.
Results In REFINE, patients with no LVEF recovery had a higher risk of sudden cardiac arrest (hazard ratio: 5.8; 95% confidence interval: 2.1 to 16.6; p = 0.001) and death (hazard ratio: 3.9; 95% confidence interval: 1.5 to 10.1; p < 0.001), independent of revascularization, baseline LVEF, and medical therapy compared with patients with recovery. Similar findings were observed in the other cohorts. LVEF reassessments beyond 6 weeks post-MI were more predictive of outcome than were earlier reassessments.
Conclusions The degree of LVEF recovery after a first MI provides important prognostic information. Patients with no recovery in LVEF after MI are at high risk of sudden cardiac arrest events and death.
Left ventricular (LV) dysfunction in the days to months following an acute myocardial infarction (MI) identifies patients at higher risk of sudden cardiac arrest (SCA) and death (1–3). Even in the era of primary percutaneous coronary intervention, baseline LV ejection fraction (EF) is an independent predictor of survival after MI (4,5).
Although improvement of LV function may occur early after MI due to recuperation of hibernating myocardium or remodeling, the degree of long-term LV recovery is tempered by adverse LV remodeling due to myocyte death and hypertrophy, inflammation, and fibrosis (5–8). The extent of this adverse cardiac remodeling is related to the development of heart failure and is associated with increased morbidity and mortality (9,10).
Mechanisms of death after MI are both time-dependent and multifactorial. During the very early period after MI, sudden death is typically related to free wall rupture or ischemia that provokes the development of lethal ventricular arrhythmias. Later, sudden death from SCA is more frequent (11,12). SCA is linked to the development of fibrosis and scar and the activation of neurohumoral pathways that may result in progressive LV dysfunction (13). These changes result in altered electrical heterogeneity, temporal dispersion of repolarization, and other factors that predispose the MI survivor to SCA (14,15).
Whereas many patients, who are treated with aggressive revascularization and medical therapy following MI, will have an improved LV systolic function, up to 50% of patients do not demonstrate improvement in LVEF several months after their index MI (16,17). The prognostic significance of the heterogeneity in LVEF change is not fully understood, yet may be important in terms of risk assessment and the routine management of post-MI patients.
The objective of the current study was not to explore LVEF cutoffs for implantable cardioverter-defibrillator (ICD) candidacy but rather to assess the prognostic value in LVEF change post-MI. We assessed the prognostic utility of LVEF recovery in 3 independent MI cohorts (REFINE [Risk Estimation Following Infarction Noninvasive Evaluation] (18); CARISMA [Cardiac Arrhythmia and Risk Stratification after Myocardial Infarction] (19); and, ISAR Improved Stratification of Autonomy Regulation (20)) to further study these issues.
Due to the heterogeneity of patient inclusion criteria and the variability in coding outcomes and variables among the 3 independent MI cohorts, each cohort was analyzed separately rather than as a pooled analysis.
REFINE included 322 patients with recent MI and LVEF ≤50% measured 2 to 7 days after the index MI (18). Of these, 248 (77%) were enrolled after a first MI. Participants were enrolled from 6 Canadian hospitals between May 2001 and July 2004. Patients with permanent atrial fibrillation, a ventricular paced rhythm, a previously implanted ICD, or an indication for an ICD at the time of enrollment were excluded (21). Baseline LVEF was assessed using a variety of methods (i.e., echocardiography in 48%, contrast ventriculography in 44%, or radionuclide ventriculography in 8%). The relevant ethics review committees at each enrollment site approved the study and all patients provided written informed consent. The primary endpoint was a cardiac death or resuscitated SCA. All-cause and cause-specific mortality were secondary outcomes. Patients were followed for a median of 4.0 (interquartile range [IQR]: 3.1, 4.7) years, and a blinded independent committee adjudicated all outcomes.
CARISMA enrolled 312 patients with recent MI and LVEF ≤40% measured 3 to 5 days following the index MI (19). Participants were enrolled from 10 European centers between August 2001 and November 2004. Of these, 188 (61%) were enrolled after a first MI. LVEF was assessed using echocardiography at baseline and again at 6 weeks post-MI. The primary endpoint was electrocardiography-documented SCA, adjudicated as “most probably treatable” by an ICD adjudicated by a blinded independent committee. All-cause mortality and cardiac death were secondary endpoints. Patients were followed for a median of 2.02 (interquartile range [IQR]: 1.97, 2.06) years.
The ISAR cohort consisted of 2,343 patients enrolled from 2 German centers (German Heart Centre and the Klinikum rechts der Isar, Munich) between January 1996 and March 2005, with a recent MI (i.e., within 4 weeks of enrollment), age ≤75 years old, in sinus rhythm and without a clinical indication for an ICD at time of enrollment (20). ISAR did not include an LVEF cutoff part of their enrolment criteria. Among patients enrolled in the ISAR cohort, 343 patients had undergone both an initial and repeat LVEF assessment within 3 months of index MI. Of these, 288 (84%) sustained a first MI and were included for data analysis. LVEF was assessed by LV angiography in the initial week after the index MI (median: day 0; IQR: 0, 1) and reassessed before 3 months (median: day 14; IQR: 12, 15). The primary endpoint of the study was all-cause mortality at 5 years. Secondary endpoints were cardiac and fatal SCA. An independent endpoint committee determined the mode of death. Patients were followed for a median of 4.9 years (IQR: 2.8, 5.0).
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 characteristics of the 3 LVEF groups were compared, including baseline characteristics, cardiovascular risk factors, cardiac history, index MI details, the extent of disease, and treatment of index MI. Continuous variables are presented as median and IQR. Categorical data are expressed as absolute numbers and percentages. Analysis of variance (parametric) and the Kruskal-Wallis test (nonparametric) were used to compare continuous variables, while the chi-square or Fisher exact tests were used to compare categorical variables among the 3 groups.
The ability of these LVEF recovery groups to predict the primary and secondary outcomes was assessed using Cox multivariate models from which hazard ratios (HR) and 95% confidence intervals (CI) were obtained. The time to development of clinical endpoints was graphically displayed by constructing Kaplan-Meier time-to-event curves and differences in survival were assessed using the log-rank test statistic. No events occurred prior to follow-up LVEF assessment. All analyses were performed using Stata 13.1 (Stata Corp., College Station, Texas). Two-sided p values ≤ 0.05 were considered significant.
The median age of participants was 60 years and the majority were male (85%). The characteristics of the 3 LVEF recovery groups, including baseline LVEF, were similar, as was the use of appropriate medical therapy. The overall median LVEF was 40% (IQR: 35, 45) measured 2 to 7 days post-MI and 49% (IQR: 39, 56) by 8 to 10 weeks after MI. Of the 248 subjects with a first MI, 31% had no recovery, 33% had a modest improvement, and 36% had large improvement (Table 1). Importantly, the LVEF early after MI did not reliably predict the subsequent LVEF value. For example, among the 63 patients with an initial LVEF of 45% to 50% early after MI, 19% had a subsequent LVEF ≤ 45% and 5% had a follow-up LVEF ≤ 35%.
During the average follow-up of 4 years, there were 17 (13 fatal and 4 nonfatal) SCA events. A strong linear relationship (p = 0.0005) in the rate of fatal or nonfatal SCA was observed among LVEF recovery groups: no recovery (15.6%), modest recovery (4.8%), and large recovery (1.1%) (Table 2, see REFINE and CARISMA data, and Figure 1A). The risk of SCA was 6-fold higher (unadjusted HR: 5.8; 95% CI: 2.1 to 16.6; p = 0.001) in patients with no LVEF recovery versus those with a modest or large improvement in LVEF (Table 3, see REFINE data). Adjustment for the presence of revascularization, index MI parameters (peak troponin, anterior location), and medical therapy did not significantly alter these relationships.
Figure 2 shows the initial and follow-up LVEF among patients that experienced SCA compared with those who did not have an event. There was no significant difference between initial and follow-up LVEF among patients with SCA. Interestingly, among the patients with SCA, 60% of patients had an initial LVEF > 35%.
To further explore the impact of initial LVEF on the strength of association between no LVEF recovery and SCA, initial LVEF was included in the model dichotomized as ≤35% or >35%. Outcomes stratified by LVEF ≤35% and >35% are shown in Table 4. Among those with an initial LVEF ≤35%, the rates of death and cardiac death were higher versus those with an initial LVEF >35% (Table 4, see baseline LVEF data). However, event rates were not statistically different among patients with a follow-up LVEF ≤35% versus >35% (Table 4, see follow-up LVEF data). When adjusted for initial LVEF (≤35% vs. >35%), revascularization, peak troponin, and anterior MI location, there was a 6-fold higher risk of SCA in patients with no LVEF recovery (adjusted HR: 5.9; 95% CI: 2.2 to 15.7; p = 0.0001). Similar results were observed when follow-up LVEF was evaluated (adjusted HR: 5.1, 95% CI: 1.1 to 23.7; p = 0.04). LVEF could not be reliably evaluated as a continuous variable due to the relatively small number of cardiac arrest events.
There were 19 deaths; 13 of which were categorized as cardiac. Linear trends were observed in the rates of death and cardiac death among patients with no recovery (14.3% and 10.4%), a moderate increase (4.8% and 4.8%), and a large increase in LVEF (4.6% and 1.1%), respectively (p = 0.01 and 0.03, respectively) (Table 2, see REFINE and CARISMA data, and Figures 1B and 1C).
Patients with no LVEF recovery had 4-fold higher risks of death (unadjusted HR: 3.9; 95% CI: 1.5 to 10.1; p = 0.0005) and cardiac mortality (unadjusted HR: 4.3; 95% CI: 1.3 to 14.7; p = 0.002) compared with patients with a modest or large recovery in LVEF. Similar results were observed after adjustment for important covariates (Table 3, see REFINE data).
We conducted an exploratory analysis to further assess the relationship between initial LVEF and the change in LVEF on cardiac mortality. Specifically, we compared 2 groups of patients: group A (n = 29) patients with initial LVEF of 30% to 35% with follow-up LVEF improvement to > 40%; and group B (n = 73) patients with initial LVEF of 40% to 50%, which remained between 40% and 50%. Cardiac mortality was higher among patients in group A (3.5%) than in group B (9.6%).
The median age of participants was 66 years and the majority were male (77%). The characteristics of the 3 LVEF recovery groups, including baseline LVEF, were similar, as was the use of appropriate medical therapy. The overall median EF was 34% (IQR: 28, 36) at 4 to 7 days post-MI and 37% (IQR: 30, 45) by 6 weeks after MI. Of the 188 subjects with a first MI, 47% had no recovery, 28% had a modest improvement, and 25% had large improvement (Table 1).
During the average follow-up of 4 years, there were 15 (10 fatal and 5 nonfatal) SCA events. A linear relationship (p = 0.038) in the rate of fatal or nonfatal SCA was observed among LVEF recovery groups: no recovery (12.4%), modest recovery (7.7%), and large recovery (0%) (Table 2, see REFINE and CARISMA data). The risk of SCA was 3-fold higher (unadjusted HR: 3.2; 95% CI: 1.0 to 10.2; p = 0.04) in patients with no LVEF recovery versus those with a modest or large improvement in LVEF (Figure 3A). We were unable to adjust for index MI parameters or presence of optimal medical therapy, as the CARISMA dataset did not contain this information.
There were 17 deaths; 10 of which were categorized as cardiac. The rates of cardiac death among patients with no recovery (9%), a moderate increase (3.9%), and a large increase in LVEF (0%), trended toward significance (p = 0.07) (Figure 3C and Table 2, see REFINE and CARISMA data). When stratified by the degree of LVEF recovery, cardiac mortality was significant higher in patients with no LVEF recovery (LVEF Δ ≤ 0%) compared with patients with modest or large LVEF recovery (LVEF Δ > 0%) (9% vs. 2%, respectively; p = 0.031). Patients with no LVEF recovery had almost a 5-fold higher risk of cardiac mortality (unadjusted HR: 4.7; 95% CI: 1.0 to 22.2; p = 0.05) compared with patients with a modest or large recovery in LVEF.
Similar to REFINE, baseline LVEF did not affect the strength of association between no LVEF recovery and adverse outcomes. When adjusted for initial LVEF (≤35% vs. >35%), there was a 5-fold higher risk of CV death in patients with no LVEF recovery compared with patients with modest or large LVEF recovery (adjusted HR: 4.8; 95% CI: 1.0 to 22.4; p = 0.05). Similar results were observed when adjusted for follow-up LVEF (≤35% vs. >35%) (adjusted HR: 5.1; 95% CI: 1.1 to 24.0; p = 0.04).
The median age of participants was 60 years and the majority were male (79%). The characteristics of the 3 LVEF recovery groups, including baseline LVEF, were similar, as was the use of appropriate medical therapy. The overall median EF was 42% (IQR: 35, 46) at 0 to 1 days post-MI and 48% (IQR: 40, 55) at 14 days. Of the 288 subjects with a first MI, 21% had no recovery, 41% had a modest improvement, and 38% had large improvement in LVEF. (Table 1).
During the median follow-up of 5 years, there were 34 deaths (19 cardiac, 12 noncardiac, and 3 undetermined). Of the cardiac deaths, 9 deaths were due to fatal cardiac arrest. There was no relationship observed between all-cause mortality, cardiac mortality, and sudden cardiac death (Table 2, see ISAR data); however, when stratified by the timing of repeat LVEF assessment (i.e., LVEF reassessed at ≥14 days), there was a trend toward increased all-cause and cardiac mortality for patients with no LVEF recovery (Table 2, see ISAR data, and Figure 4). With LVEF reassessments performed beyond 14 days, patients with no LVEF recovery had a 2-fold higher risk of cardiac mortality (unadjusted HR: 2.3; 95% CI: 0.9 to 6.3; p = 0.09) and an almost 4-fold higher risk of all-cause mortality (unadjusted HR: 3.6; 95% CI: 0.9 to 14.2; p = 0.07) (Table 3, see ISAR data).
The degree of improvement in LVEF over the initial 3 months after a first MI is a consistent predictor of cardiac mortality across 3 patient cohorts (REFINE, CARISMA, and ISAR). Additionally, REFINE and CARISMA reveal that lack of LVEF recovery is associated with a higher risk of SCA, independent of baseline LVEF and aggressive contemporary management, such as revascularization and medical therapy.
LVEF assessment early after MI has been shown to predict long-term cardiac morbidity and mortality (2,3,22), including in patients who have undergone revascularization and are treated with contemporary medical therapy (4,5,23). However, the data supporting the use of change in LVEF to predict outcomes is much more sparse (17,24). Parodi et al. (24) found that the survival of MI survivors with a ≥10% increase in LVEF over 6 months (8%) was over 2-fold greater than those without such an improvement (18%) over a follow-up of 5 years (p = 0.02). However, incomplete follow-up, lack of adjustment for important covariates and the single center nature of these results represent important limitations. The results of the present analysis add to the existing literature by demonstrating a consistent association between absence of LVEF recovery and increased cardiac mortality and SCA in 3 independent post-MI population cohorts from Europe and Canada (REFINE, CARISMA, and ISAR), even among patients with initial LVEF >35%.
Persistent LV dysfunction
Previous studies have shown that a sizeable proportion of patients sustaining myocardial infarction do not demonstrate a significant increase in LVEF over the initial few months after an acute MI despite contemporary management (17,24). The present analysis is consistent with the existing literature: 31% in the REFINE cohort, 47% in CARISMA, and 21% in ISAR had a lack of improvement in their LVEF. A lack of improvement translated into increased risk of SCA and cardiac mortality.
Timing of LV reassessment
Whereas the REFINE and CARISMA cohorts demonstrated a significant association between LVEF recovery and clinical outcomes, the ISAR cohort demonstrated a trend toward increased all-cause and cardiac mortality for patients with adverse remodeling. The lack of statistical significance could be due to the timing of repeat LV assessment. The association between LVEF recovery and outcome only became apparent in the subgroup of the ISAR cohort that underwent repeat LV assessment beyond 14 days (median: 16 days; IQR: 15, 19). Of these patients, only 5 underwent LV reassessment beyond 4 weeks; and 2 patients underwent reassessment beyond 6 weeks. In comparison, the CARISMA cohort reassessed LV function after a minimum of 6 weeks following index MI and the REFINE cohort reassessed LV function between 8 to 10 weeks following index MI.
Reassessment of LVEF too early following MI (i.e., less than 6 weeks following index MI) may not adequately discriminate between myocardial stunning and adverse LV remodeling; whereas, LVEF reassessment beyond 6 weeks more likely reflects presence or absence of adverse remodeling (17,25). Indeed, the inability to discriminate adverse remodeling with early LVEF assessments may allude to the lack of mortality benefit seen in the trials of early ICD implantation post-MI (26,27).
Initial LVEF versus no LVEF recovery: Which is the more robust prognostic marker?
Both the absence of LV recovery and an initial LVEF ≤35% are independent predictors of adverse events. Our exploratory analysis using the REFINE cohort data sought to assess the relative predictive strengths of initial LVEF and no LVEF recovery to determine cardiac mortality. Interestingly, patients with lower initial LVEF values that “recovered” (group A) had a lower risk of cardiac mortality (3.5%) compared with patients in group B who had a less severe LVEF at the start, but did not experience LVEF improvement (9.6%). It is important to highlight that this exploratory analysis is limited by the small number of events and warrants further study.
Determinants of LV improvement
Given likely differences in LVEF recovery in patients with an initial MI event versus multiple prior MI events, we confined our analysis to patients with a first MI (4). Prior studies have suggested a relationship between LVEF recovery and a variety of clinical variables including the time from symptom onset to successful reperfusion, anterior location of infarction, female sex, and the rise in cardiac enzymes (24,28,29). However, other studies have failed to confirm the relationship between these variables and the change in LVEF after MI (4,17,24,29,30). Whereas peak troponin or other cardiac enzymes could be assumed to represent the most reliable predictor of LV remodeling given the link between enzymatic infarct size and mortality (31,32), the relationships among infarct size, peak troponin, and change in LVEF are inconsistent (33–35).
Indeed, the predictors of LVEF change are not consistent among the cohorts included in the present study. ISAR revealed an association between LVEF change and peak creatinine kinase, anterior location of infarction, or presence of beta-blocker therapy; however, the same association was not seen in REFINE. The reasons for this are unclear. As noted, the relationship between initial cardiac biomarker and change in LVEF has not been consistently found in prior studies. Furthermore, we did not collect serial cardiac enzyme data and the area under the cardiac enzyme curve may provide additional prognostic information (36).
Repeat LVEF assessment
Baseline LVEF was not different among the patients with no, modest, or large subsequent increases in LVEF. The fact that a sizeable proportion of the patients assessed had no improvement or a decline in LVEF after MI illustrates that despite optimal contemporary post-MI management, a subset of patients will undergo unfavorable LV remodeling (16,28). The data from the present analysis would also indicate that reliance on the baseline LVEF as a measure of long-term LVEF is ill-advised given that a sizable minority of patients with initial LVEF values of 45% to 50% early on had follow-up LVEF values ≤40% when reassessed at 2 to 3 months post-MI.
The similar baseline characteristics among these groups implies that clinical factors may be of limited value in aiding the clinician in deciding whether a patient does or does not require a follow-up LVEF assessment after MI (37–40). The 2013 American College of Cardiology/American Heart Association Guidelines for Management of Patients with ST-Elevation Myocardial Infarction recommend that LV function be reevaluated ≥40 days following MI with significant LV systolic dysfunction (LVEF ≤40%) to assess potential need for implantable cardioverter defibrillator therapy (41). Yet, even patients with initial LVEF values of 45% to 50% early after MI may have significant residual LV dysfunction.
Clinical implications and direction for future studies
The current study supports LVEF change as a prognostic marker of adverse outcomes and reinforces current guideline recommendations to reassess LVEF post-MI (41,42). Despite these recommendations, suboptimal LVEF reassessment rates (<50%) are observed in real-world practice (43,44). A proportion of patients experience further deterioration in LVEF upon follow-up, which may qualify them for ICD implantation. Importantly, low rates of LVEF reassessment were associated with low rates of appropriate ICD implantation (44). Additional study into the processes affecting the low rates of LVEF reassessment is required, as these suboptimal LVEF reassessment rates may result in missed care opportunities.
The prognostic value of change in LVEF post-MI as a marker of adverse events also warrants further study. Despite baseline LVEF occupying a central role in current guideline recommendations for ICD implantation post-MI, LVEF has limitations as a sole risk-stratification technique for SCA (45). For example, among patients in the REFINE cohort who experienced SCA, 60% of patients who experienced SCA had an initial LVEF >35% (Figure 2). Our findings are consistent with the literature; in cohort of 2,100 post-MI patients, two-thirds of the SCA events occurred in patients with an LVEF >35% despite a lower incidence of SCA compared to patients with LVEF ≤35% (46). Future studies are required to develop and validate more precise risk stratification tools for SCA post-MI, which may include the combination of LVEF change, baseline LVEF and other emerging or existing SCA risk stratification techniques (such as late gadolinium enhancement on cardiac magnetic resonance imaging, microvolt T-wave alternans, and abnormal heart rate variability) (45).
These results must be interpreted in the context of some inherent study limitations, which include a post hoc analysis. There were several methods used for the LVEF measurement, which may limit precision of LVEF quantification and comparison. Additionally, the LVEF recovery was treated categorically instead of as a continuum. However, the categories chosen allow for simplicity in clinical use, and as shown by the present study, carry clinical relevance. Another limitation was the relatively low number of outcome events, which resulted in imprecise estimates of the HR with wide CI. Finally, the time to LVEF reassessment post-MI varied across the 3 MI cohorts (from 2 to 8 weeks). Along with the heterogeneity in the inclusion criteria of the MI cohorts, a pooled cohort analysis was not deemed suitable.
Our study shows that in patients with first presentation MI, the absence of LVEF recovery is independently associated with increased risk of serious events in follow-up, including a nearly 6-fold risk of nonfatal and fatal cardiac arrest, or over a 4-fold risk of all-cause mortality. However, the prognostic ability of LVEF recovery may be limited by the timing of reassessment. LVEF recovery based on repeat measurements beyond 6 weeks were more strongly predictive of outcome, compared with LVEF recovery based on repeat EF measurements at 2 weeks.
COMPETENCY IN MEDICAL KNOWLEDGE: Despite revascularization and optimal contemporary medical therapy post-MI, a sizeable proportion of patients will undergo unfavorable LV remodeling. Lack of LVEF recovery provides additional insight into post-MI risk stratification and is associated with increased risk of adverse events independent of initial absolute EF post-MI.
TRANSLATIONAL OUTLOOK: Prospective studies are needed to further validate the prognostic value of EF change during the months following MI and to identify factors associated with LV remodeling. The development and validation of a clinical prediction model would help target at-risk populations for comprehensive and aggressive antiremodeling strategies.
Dr. Chew is supported by an Arthur J.E. Child Cardiology Fellowship, and is a member of the Cardiac Arrhythmia Network of Canada (CANet) HQP Association for Trainees (CHAT). Dr. Exner has received consulting fees from GE Healthcare, Medtronic, Abbott Medical, Boston Scientific; and has equity in Analytics for Life. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- confidence interval
- ejection fraction
- hazard ratio
- implantable cardioverter-defibrillator
- interquartile range
- left ventricle
- myocardial infarction
- sudden cardiac arrest
- Received April 28, 2017.
- Revision received December 22, 2017.
- Accepted December 28, 2017.
- 2018 American College of Cardiology Foundation
- Hammermeister K.E.,
- DeRouen T.A.,
- Dodge H.T.
- White H.D.,
- Norris R.M.,
- Brown M.A.,
- Brandt P.W.,
- Whitlock R.M.,
- Wild C.J.
- Solomon S.D.,
- Skali H.,
- Anavekar N.S.,
- et al.
- Braunwald E.,
- Kloner R.A.
- Camici P.G.,
- Prasad S.K.,
- Rimoldi O.E.
- Gaudron P.,
- Eilles C.,
- Kugler I.,
- Ertl G.
- Zipes D.P.,
- Wellens H.J.
- Stevenson W.G.,
- Khan H.,
- Sager P.,
- et al.
- Exner D.V.,
- Kavanagh K.M.,
- Slawnych M.P.,
- et al.,
- for the REFINE Investigators
- Gregoratos G.,
- Abrams J.,
- Epstein A.E.,
- et al.
- Braunwald E.,
- Rutherford J.D.
- Ndrepepa G.,
- Mehilli J.,
- Martinoff S.,
- Schwaiger M.,
- Schomig A.,
- Kastrati A.
- Funaro S.,
- La Torre G.,
- Madonna M.,
- et al.,
- for the AMICI Investigators
- Newby L.K.,
- Christenson R.H.,
- Ohman E.M.,
- et al.,
- for the GUSTO-IIa Investigators
- Hallen J.,
- Jensen J.K.,
- Fagerland M.W.,
- Jaffe A.S.,
- Atar D.
- Pfeffer M.A.,
- Braunwald E.
- Anderson J.L.,
- Adams C.D.,
- Antman E.M.,
- et al.
- Hamm C.W.,
- Bassand J.P.,
- Agewall S.,
- et al.
- Jneid H.,
- Anderson J.L.,
- Wright R.S.,
- et al.
- O'Gara P.T.,
- Kushner F.G.,
- Ascheim D.D.,
- et al.
- Bennett M.,
- Parkash R.,
- Nery P.,
- et al.
- Miller A.L.,
- Gosch K.,
- Daugherty S.L.,
- et al.