The impact of high‐intensity interval training on ventricular remodeling in patients with a recent acute myocardial infarction—A randomized training intervention pilot study

Abstract Background Aerobic exercise training is associated with beneficial ventricular remodeling and an improvement in cardiac biomarkers in chronic stable heart failure. High‐intensity interval training (HIIT) is a time‐efficient method to improve V˙O2peak in stable coronary heart disease patients. This pilot study aimed to compare the effect of HIIT on ventricular remodeling in patients with a recent acute myocardial infarction (AMI). Methods Nineteen post‐AMI patients were randomized to either HIIT (n = 9) or usual care (n = 10). A cardiopulmonary exercise test (CPET), transthoracic echocardiography, and cardiac biomarker assessment (ie, N‐terminal pro B‐type natriuretic peptide levels and G protein‐coupled receptor kinase 2 expression) were performed before and after a 12‐week training intervention. CPET parameters including oxygen uptake efficiency slope (OUES) and V˙O2 at the first ventilatory threshold (V˙O2 VT1) were calculated. left ventricular (LV) structural and functional echocardiographic parameters including myocardial strain imaging were assessed. Results V˙O2peak and OUES improved solely in the HIIT group (P < .05 for group/time, respectively). There was a significant training effect for the improvement of peak work load in both groups (P < .05). O2 pulse and V˙O2 at VT1 both improved only in the HIIT group (P < .05 for time, no interaction). HIIT improved radial strain and pulsed‐wave tissue Doppler imaging derived e′ (P < .05 for time, no interaction). Cardiac biomarkers did not change in either group. Conclusions In post‐AMI patients, HIIT lead to significant improvements in prognostic CPET parameters compared to usual care. HIIT was associated with favorable ventricular remodeling regarding certain echocardiographic parameters of LV function.


| INTRODUCTION
Acute myocardial infarction (AMI) can induce changes in left ventricular (LV) topography (ie, ventricular remodeling) and is a major contributor in the development of heart failure despite advances in coronary revascularization and optimal medical therapy. 1 Myocardial strain imaging using speckle-tracking echocardiography allows quantification of regional and global LV function and has been increasingly implemented in clinical practice. 2 It is more sensitive for the detection of subclinical LV changes as compared to standard LV ejection fraction (LVEF) measurement. 3 The best evaluated parameter, global longitudinal strain (GLS) is superior to LVEF in the prediction of prognosis and cardiac remodeling after AMI. [4][5][6] Among others, cardiac β-adrenergic receptor (β-AR) signal dysregulation represents a hallmark abnormality potentially leading to LV remodeling post-AMI and progression to heart failure. β-AR kinase (GRK2) is the most abundant G protein-coupled receptor kinase expressed in the heart. 7 Importantly, abnormalities of β-AR signaling in the failing heart, including GRK2 over-expression, are mirrored in circulating white blood cells (ie, lymphocytes) and correlate with severity of LV dysfunction. 8 Therefore, GRK2 provides potential as a biomarker of cardiac dysfunction. 9 Exercise-based secondary prevention programs have confirmed improvements in mortality and morbidity in patients with stable coronary heart disease (CHD) and after AMI, respectively. 10,11 The importance of starting aerobic exercise training early post-AMI and the beneficial effects on LV remodeling have been emphasized in a recent meta-analysis. 12 Furthermore, aerobic exercise training has been shown to be associated with a lowering of GRK2 expression and to predict outcomes in patients with chronic ischemic heart failure in a prospective study. 13 Highintensity interval training (HIIT) is more effective at improving _ VO 2peak and can be performed safely compared to the more established moderate-intensity continuous exercise training (MICET) in stable CHD patients. [14][15][16] However, most prior studies included predominantly stable patients. To the best of our knowledge, the effect of HIIT on cardiac remodeling including advanced echocardiography (ie, myocardial strain imaging) and GRK2 expression has not yet been studied in patients with a recent AMI. This

| Study design and measurement
Baseline clinical assessment (ie, medical history, physical examination, and anthropometric measurements), blood analysis, transthoracic echocardiography, and CPET were performed at baseline and after completion of the program (for more detailed information on study design and measurement see section S2 of the Supporting Information).

| Maximal CPET
Maximal CPET was performed on a cycle ergometer (Ergoline 800S, Bitz, Germany) according to the recommendations of the American Heart Association, and as previously published (for more detailed information see section S3 of the Supporting Information). 20-22

| Transthoracic echocardiography
Standard transthoracic 2D echocardiography was performed on a Vivid 9 cardiac ultrasound system with a 7.5-MHz transducer (GE Medical system, New Jersey). All echocardiographic images were obtained by two cardiology fellows using standard tomographic views.
All data were stored on an external hard-drive and analyzed offline on a commercially available workstation (EchoPAC, GE Healthcare) by a cardiology fellow and checked by single experienced cardiologist with several years of expertise. Traditional echocardiographic parameters of LV dimension, and systolic and diastolic function were assessed based on the most recent recommendations. 23,24 2.5 | 2D speckle-tracking strain analysis Peak systolic LV longitudinal strain and strain rates were assessed using standard 2D apical four-chamber, two-chamber, and threechamber view using speckle-tracking analysis. 25

| G protein-coupled receptor kinase 2
The expression of G protein-coupled receptor kinase 2 (GRK2) was measured by Western blotting (for more detailed information on GRK2 evaluation see section S4 of the Supporting Information).

| Exercise training intervention
Patients were randomized to either a 12-week structured exercise training program including two weekly supervised HIIT sessions or a usual care group. An additional resistance training (RT) was performed following each HIIT session. All trainings were center-based under supervision of an experienced kinesiologist. The HIIT training protocol was recently described by Guiraud et al. 21 Following a 5-minute warm-up at 30% of peak work load obtained at the CPET, patients performed two to three sets of 6 to 8 minutes with repeated bouts of 15 to 30 seconds at 100% of peak work load alternated by 15 to 30 seconds of passive recovery.
The targeted Borg rating of perceived exertion (RPE) was set at 15 during the HIIT bouts. The sets were separated by a 5-minute active recovery phase at 30% of peak work load. The training session was terminated by a 5-minute cool-down phase at 30% of peak work load. 26

| Statistical analyses
Data are presented as mean ± SD for continuous variables, while frequencies and percentages are presented for categorical variables.
Baseline characteristics were compared between the two groups using Student t test in case of continuous variables and categorical variables were compared using chi-square test. Repeated measures ANOVA models were used to study the CPET and echocardiographic parameters across time and between groups. Models with time, group, and group × time interaction as independent variables were used. The group × time interaction was the main focus of the analysis as it tested the difference in the change (post-pre) between the two groups. As a measure of effect size to evaluate the strength of the intervention effect (HIIT) vs usual care, the Hedge's g calculated by the formula below was presented. where

| Clinical characteristics
A total of 19 were included in the final analysis (HIIT: n = 9, usual care group: n = 10). Baseline clinical characteristics of CHD patients with a recent AMI, either randomized to the HIIT or usual care group are summarized in Table 1. There were no significant differences between the groups with regards to demographic data, event details and baseline medication except for a lower number of patients on inhibitors of the renin angiotensin aldosterone system (ACE inhibitor or angiotensin receptor blocker) in the HIIT group (P < .05).

| Echocardiographic parameters
Echocardiographic parameters of LV geometry, systolic and diastolic function in both groups are summarized in Table 3 for a more detailed description of the results of echocardiographic parameters we refer to section S6 of the supplemental text).   MICET in a population of CHD patients who reported the most pronounced improvements with HIIT after 7 to 12 weeks of training. 16,32 The finding that _ VO 2 at the VT1 improved only in the HIIT group indicates an improvement in muscular function as a result of structured exercise training and is plausible. 33 T A B L E 3 (Continued)

| The impact of HIIT on echocardiographic parameters of cardiac remodeling
To the best of our knowledge, our pilot study for the first time shows beneficial effects of a HIIT program on cardiac remodeling in patients with a recent AMI. Existing literature has shown that aerobic exercise training starting early after AMI has no detrimental effects and even reverses ventricular remodeling in post-AMI LV dysfunction. 12,34,35 Additional data suggest a more favorable cardiac remodeling after HIIT as compared to the more established MICET or a control group in chronic stable post-AMI patients with heart failure. 36 However, the effects of HIIT in patients with a recent AMI and potential acute transient LV dysfunction have never been examined.
First of all, patients in our cohort were on optimal and stable medical therapy. The primary explanation for the lower number of patients on inhibitors of the renin angiotensin aldosterone system in the HIIT group is the lower number of STEMI patients in this group and the absence of LV dysfunction requiring this medication in the whole cohort before inclusion into the study. Importantly, after completion of the program all patients in our study except for one had a normal LVEF, left ventricular mass and volumes (indexed by body surface area), and no higher degree diastolic dysfunction (ie, diastolic dysfunction > grade I) based on recent recommendations. 23,24 In our pilot study, all these "conventional" parameters remained stable with HIIT, which is in line with other studies that employed lower training intensities 12,35 Furthermore, e 0 septal as a parameter of diastolic function showed a significant improvement with training only in the HIIT group. However, there were no significant changes with e 0 lateral and E/e 0 , respectively. Thus far, HIIT has been shown to improve e 0 in post-AMI heart failure patients only. 36  show an improvement in radial strain after HIIT in post-AMI patients. 40 This contrasts with studies that report a decrease of LV systolic (including strain analysis) and diastolic functional parameters. 41,42 However, these findings in healthy subjects and athletes seem to be transient and particularly after prolonged and strenuous exercise and are discussed controversial.

| The impact of HIIT on cardiac biomarkers
Based on the CPET and echocardiographic findings in our study, the statement that no patient in our study developed heart failure within the first months of a first AMI is of utmost importance. This is confirmed in normal NT-pro BNP levels in our patient cohort. On the one hand this mirrors optimized treating strategies in recent decades, but it also explains that only insignificant changes in cardiac biomarkers (particularly GRK2 expression) were detected in this first exploration of CHD patients who recently suffered an AMI undergoing HIIT.

| Limitations
Our findings have to be interpreted in the context of numerous limitations. First of all, the sample size in the present pilot study was small with inclusion of predominantly male patients at a single tertiary institution. The initial power calculation to randomize 10 patients to each arm (20 total) calculated to have 80% power to demonstrate a significant reduction of GRK2 expression with HIIT in this population was based on a prospective exercise training study in patients with chronic heart failure. 13 Moreover, data of one patient in the HIIT group were F I G U R E 1 GRK2 protein expression pre-and post-training with either usual care or HIIT. Top: typical Western blots are shown. Bottom: Paired individual GRK2 expressions (normalized to GAPDH and to 100% in pre-training) are shown in each patient. In red, mean ± SD of the changes in GRK2 expression for each type of exercise is shown, n = 8 in MICET and n = 9 in HIIT. GRK2, G protein-coupled receptor kinase 2; HIIT, high-intensity interval training; MICET, moderate-intensity continuous exercise training not available for GRK2 expression. Only eight patients in the HIIT group underwent an echocardiographic examination before and after completion of the training intervention. For the most relevant advanced echocardiographic parameters (ie, GLS) there were no data of at least one more patient in the whole cohort due to insufficient tracing and missing data.
In summary, the finding that no patient developed heart failure in our study may emphasize improved treating strategies over the last decades in this population (ie, revascularization, medication) on one hand. On the other hand, the fact that no adverse event occurred during the study together with the reported findings may indicate that HIIT is safe in this specific population. Future studies may-be applied to a sicker cohort (ie, patients with confirmed LV dysfunction post-AMI).

| CONCLUSIONS
In patients with a recent AMI without LV dysfunction, HIIT leads to significant improvements regarding prognostic CPET parameters ( _ VO 2peak , OUES) compared to a usual care group. HIIT overall exhibited a more favorable cardiac remodeling with regard to echocardiographic parameters of LV function.

ACKNOWLEDGMENTS
The study was funded by the Montreal Heart Institute Foundation and the EPIC Center Foundation.