Abstract
ackground:Left ventricular (LV) mechanics assessed by speckle-tracking echocardio- graphy provides sensitive markers of cardiac adaptation to exercise. Different training modalities—endurance, high-intensity interval training (HIIT), and acute exercise tests— impose distinct hemodynamic loads, yet their comparative effects on LV deformation remain unclear. Importantly, acute and chronic endurance exposures may elicit divergent myocardial responses that must be interpreted separately.Methods:A systematic search of PubMed, Scopus, and EMBASE (through September 2025) identified studies evaluating LV mechanics in response to endurance, HIIT, or acute exercise among healthy or recreationally active individuals. Echocardiographic parameters of strain and torsion were extracted, and methodological quality was appraised using the NIH Quality Assessment Tool.Results: Twenty-three studies (859 participants) met inclusion criteria. Acute prolonged endurance exercise—particularly marathon and ultra-endurance events—was associated with tran- sient, fully reversible reductions in global longitudinal, circumferential, and radial strain and torsion, despite preserved ejection fraction, reflecting short-term myocardial fatigue rather than maladaptive remodeling. In contrast, chronic endurance training maintained or improved LV mechanics without evidence of dysfunction, while HIIT interventions consistently enhanced LV systolic strain and rotational indices across diverse age groups and sexes, reflecting improved contractile efficiency and physiological remodeling. Acute exercise produced heterogeneous, load-dependent strain responses, with isometric stress increasing
preserved ejection fraction, reflecting short-term myocardial fatigue rather than maladaptive remodeling. In contrast, chronic endurance training maintained or improved LV mechanics without evidence of dysfunction, while HIIT interventions consistently enhanced LV systolic strain and rotational indices across diverse age groups and sexes, reflecting improved contractile efficiency and physiological remodeling. Acute exercise produced heterogeneous, load-dependent strain responses, with isometric stress increasing regional strain and maximal exertion inducing temporary global reductions. Between-study heterogeneity was moderate, methodological quality generally good, and small-study effects varied by modality, being most evident in endurance studies, borderline for HIIT, and limited for acute tests due to sample size.Conclusions:Acute endurance exercise produces transient, reversible LV deformation changes, whereas chronic endurance training preserves mechanical efficiency. HIIT reliably enhances systolic strain and torsional mechanics, and acute exercise elicits variable but physiologically meaningful responses. These findings clarify that transient post-race strain reductions reflect physiological fatigue, not chronic maladaptation, and underscore the modality-specific nature of myocardial adaptation to exercise. Keywords:left ventricular mechanics; speckle-tracking echocardiography; endurance training; high-intensity interval training; acute exercise; strain; torsion J. Clin. Med.2025,14, 8210 https://doi.org/10.3390/jcm14228210
J. Clin. Med.2025,14, 8210 2 of 26 1. Introduction Regular physical exercise triggers profound adaptations in cardiac structure and func- tion, which have been the subject of intense investigation for decades [1,2]. Traditionally, studies of the “athlete’s heart” have focused on gross parameters such as ventricular chamber size, wall thickness, and left ventricular ejection fraction (LVEF) [3,4]. Yet, these conventional echocardiographic indices often remain within normal limits, even in the presence of subtle myocardial remodeling or dysfunction. In recent years, myocardial de- formation imaging—particularly speckle-tracking echocardiography (STE)—has emerged as a more sensitive tool to detect early or subclinical changes in left ventricular (LV) performance [5,6]. Left ventricular global longitudinal strain (LV–GLS) is known to decline before overt changes in LVEF occur and is prognostically informative in various cardiovascular disease settings [7,8]. A recent meta-analysis examined how exercise interventions influence LV– GLS and found that, while significant improvements were evident in populations with cardiovascular disease, in healthy or low-risk individuals the effects were more variable and often modest [9]. This inconsistency underscores the need for a more comprehensive assessment of how different forms of exercise influence LV mechanics under diverse loading and adaptation conditions. From a physiological perspective, LV mechanics include longitudinal deformation, circumferential shortening, radial thickening, and twist–untwist (torsion), all of which contribute to efficient systolic ejection and diastolic recoil [10]. Exercise imposes complex demands on preload, afterload, and contractility, and the adaptation of each mechanical component may depend on the intensity, duration, and nature of the training stimulus [11]. Earlier studies in healthy individuals suggest that exercise training may induce regional heterogeneity in LV systolic function, possibly mediated by right ventricular adaptations and ventricular interdependence [12,13]. In interpreting exercise-induced changes in myocardial strain, it is essential to dis- tinguish between acute endurance bouts and chronic training adaptations. In athletes or trained populations, acute endurance exposure—such as marathon or ultramarathon events—can lead to transient “myocardial fatigue,” characterized by fully reversible re- ductions in strain and delayed untwisting without evidence of lasting dysfunction [14]. Conversely, chronic endurance training promotes physiological (eccentric) remodeling and preserves or enhances strain indices over time [15].
between acute endurance bouts and chronic training adaptations. In athletes or trained populations, acute endurance exposure—such as marathon or ultramarathon events—can lead to transient “myocardial fatigue,” characterized by fully reversible re- ductions in strain and delayed untwisting without evidence of lasting dysfunction [14]. Conversely, chronic endurance training promotes physiological (eccentric) remodeling and preserves or enhances strain indices over time [15]. High-intensity interval training (HIIT) imposes repeated stress–recovery cycles that may promote more efficient contractile adaptations and enhance strain or torsional reserve [16]. Controlled acute tests (e.g., iso- metric or short maximal efforts) also provide insight into myocardial reserve and stress responsiveness [17]. Despite increasing use of strain imaging in exercise physiology, few studies have di- rectly compared endurance, HIIT, and acute exercise models. Many have focused on single modalities or specific populations, making cross-study comparisons difficult. Therefore, in this systematic review and meta-synthesis, we evaluate studies assessing LV mechanics— both conventional and deformation parameters—across these three exercise modalities. Our objective is to delineate the distinct mechanical signatures associated with endurance training, HIIT, and acute exercise, and to clarify the physiological mechanisms underlying these adaptations. We specifically aim to distinguish transient, load-dependent alterations following acute endurance exercise from chronic remodeling processes induced by sus- tained training exposure. We hypothesize that acute endurance bouts are more frequently associated with short-term, reversible reductions in deformation, HIIT with consistent en- hancement of strain and torsional efficiency, and chronic endurance training with preserved or improved mechanical performance over time.
J. Clin. Med.2025,14, 8210 3 of 26 2. Materials and Methods This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [18] (Supplemen- tary Materials S1) and was prospectively registered in the INPLASY database (ID: IN- PLASY2025100002) on 2 October 2025. The full record (https://inplasy.com/inplasy-2025 -10-0002/ 2.1. Search Strategy We conducted a comprehensive literature search to identify studies evaluating the effects of different exercise training modalities on LV mechanics assessed by STE. The search was performed in PubMed, Scopus, and EMBASE databases from inception through September 2025, with no language restrictions. The search strategy combined Medical Subject Headings (MeSH) and free-text terms, including: “endurance training,” “high-intensity interval training,” “acute exercise,” “left ventricular mechanics,” “speckle- tracking,” “strain,” “torsion,” and “echocardiography.” Boolean operators (“AND,” “OR”) were applied to maximize sensitivity. In addition, the reference lists of relevant systematic reviews and included articles were screened manually to identify additional eligible studies. Although no language restrictions were applied, all studies retrieved and included were published in English; therefore, no translation procedures were required. 2.2. Eligibility Criteria Studies were considered eligible if they were original research articles published in peer-reviewed journals and evaluated the effects of physical training or acute exercise on LV mechanics using STE. We included studies that investigated healthy participants across a range of activity levels, from competitive athletes and recreationally active in- dividuals to sedentary subjects undergoing structured training interventions. Eligible exercise exposures comprised endurance training, HIIT, or acute physiological tests such as isometric exercise, short maximal efforts, or cardiopulmonary exercise testing. To be included, studies were required to report at least one echocardiographic measure of LV deformation, such as GLS, global circumferential strain (GCS), global radial strain (GRS), or torsional parameters, either before and after the intervention or between different training phases. Conventional echocardiographic indices and biomarkers were also extracted when available. Exclusion criteria included case reports, conference abstracts without sufficient data, narrative reviews, and editorials. Previous meta-analyses were also excluded to avoid duplication of aggregated data. We also excluded studies conducted in populations with overt cardiovascular disease, structural cardiac abnormalities,
either before and after the intervention or between different training phases. Conventional echocardiographic indices and biomarkers were also extracted when available. Exclusion criteria included case reports, conference abstracts without sufficient data, narrative reviews, and editorials. Previous meta-analyses were also excluded to avoid duplication of aggregated data. We also excluded studies conducted in populations with overt cardiovascular disease, structural cardiac abnormalities, or other clinical conditions that might independently alter LV mechanics. When duplicate reports derived from the same cohort were identified, the most complete or most recent study was retained. 2.3. Study Selection and Data Extraction Two investigators independently screened titles and abstracts retrieved from the initial search to exclude irrelevant records. Full texts were then assessed for eligibility. Discrep- ancies were resolved by consensus with a third reviewer. Data extraction was performed independently by three experienced cardiologists through September 2025, using a stan- dardized form. Extracted data included: study author, year of publication, country, sample size, age and sex distribution, population characteristics (athletes vs. non-athletes), exercise modality and duration, study design, echocardiographic platform, imaging parameters (GLS, GCS, GRS, torsion, twist/untwist rates), conventional echocardiographic indices, and biomarker data when available. Data were cross-checked for accuracy and summarized in structured tables.
J. Clin. Med.2025,14, 8210 4 of 26 2.4. Risk of Bias Assessment The methodological quality of the included studies was assessed using the National Institutes of Health (NIH) Quality Assessment Tool for Observational Cohort and Cross- Sectional Studies [19]. This instrument evaluates 14 domains, including clarity of the research question, definition of the study population, participation rate, exposure and outcome measures, adequacy of statistical analyses, and length of follow-up for intervention studies. Each study was independently rated by the three reviewers who performed data extraction and classified as good, fair, or poor quality according to NIH guidelines. 2.5. Statistical Analysis Extracted continuous data were expressed as median and interquartile range when reported in that format by the original studies. Where both pre- and post-exercise or pre- and post-training values were available, mean differences (∆) were calculated to quantify within-study change. When only baseline and follow-up means with standard deviations (SDs) were provided,∆= Mean_post−Mean_pre, and the SD of the mean difference (SD∆) was derived using the formula: SD∆= √ (SD_pre 2 + SD_post 2 − 2×r×SD_pre×SD_post), assuming a pre–post correlation coefficient (r) of 0.70 based on prior echocardiographic training interventions. Sensitivity analyses were performed using alternative r values (0.5 and 0.9) to confirm robustness. When studies reported only standard errors (SEs), SDs were back-calculated (SD = SE× √n). If dispersion data were missing, they were imputed from group medians and interquartile ranges using established statistical conversion methods. In accordance with standard speckle-tracking echocardiography practice, more neg- ative GLS and GCS values represent greater myocardial deformation (i.e., functionally “better”). To ensure consistency across studies and pooled analyses, all∆values were normalized to a unified directionality: positive∆indicates an increase (i.e., more nega- tive/improved strain for GLS and GCS), while negative∆reflects a reduction (less nega- tive/worsened strain). These∆values were subsequently grouped by exercise modality: endurance training, HIIT, and acute exercise testing. Normality of data distribution was verified using the Shapiro–Wilk test. Because all variables were non-normally distributed, between-group comparisons among the three exercise modalities were performed using the Kruskal–Wallis test, followed by Dunn’s post hoc correction for multiple pairwise comparisons. Statistical significance was set atp<
tive/worsened strain). These∆values were subsequently grouped by exercise modality: endurance training, HIIT, and acute exercise testing. Normality of data distribution was verified using the Shapiro–Wilk test. Because all variables were non-normally distributed, between-group comparisons among the three exercise modalities were performed using the Kruskal–Wallis test, followed by Dunn’s post hoc correction for multiple pairwise comparisons. Statistical significance was set atp< 0.05. Outcomes were further categorized as hemodynamic (heart rate, systolic and diastolic blood pressure), conventional echocardiographic (LV chamber dimensions, mass, ejection fraction, stroke volume), or myocardial deformation parameters (global longitudinal, cir- cumferential, and radial strain, as well as rotational and torsional indices). For quantitative synthesis, pooled∆values were computed as weighted mean differences (WMDs) with corresponding 95% confidence intervals (CIs) using a random-effects model (DerSimonian– Laird method). Key pooled outcomes (e.g.,∆GLS,∆GCS,∆torsion) are reported with their 95% CIs and I 2 statistics in Section Between-study heterogeneity was quantified using Cochran’s Q and expressed as I 2 , calculated as 100×(Q−df)/Q, representing the proportion of total variance attributable to inter-study differences rather than sampling error. I 2 values of approximately 25%, 50%, and 75% were interpreted as low, moderate, and high heterogeneity, respectively. WhereI 2 > 60%or model convergence was unstable, pooled values were summarized descriptively without statistical inference. Sensitivity analyses were conducted by examining the influence of study weighting and by qualitatively comparing directionality across modalities. Publication bias was assessed visually using funnel plots and quantitatively using Egger’s regression test for small-study effects. The regression intercepts (Egger’s test) were estimated separately for
J. Clin. Med.2025,14, 8210 5 of 26 each exercise modality group (acute testing, endurance, and HIIT), with intercept values close to zero andp> 0.05 suggesting no significant small-study bias. Results were synthesized narratively and quantitatively, with pooled effect sizes and I 2 values summarized in tables and forest plots. All computations were performed using Comprehensive Meta-Analysis v3.0 (Biostat, Englewood, NJ, USA), and descriptive plots were generated using GraphPad Prism v10. 3. Results 3.1. Study Selection The initial research performed in PubMed, Scopus, and Embase databases identified 147 studies evaluating the effects of different exercise training modalities on LV mechanics in athletes. Fifteen studies (10.2%) were removed as duplicates, and 99 (67.3%) were excluded on the basis of the prespecified exclusion criteria. The remaining 33 studies (22.4%) were assessed for eligibility. Of these, 4 (2.7%) were excluded due to incomplete clinical data and 6 (4.1%) due to incomplete STE data. Accordingly, 23 studies(15.6%) [20–42] were included in this systematic review and meta-analysis, totaling 859 participants across diverse age groups and training backgrounds (Figure). Figure 1.Flowchart illustrating the process of study identification, screening, eligibility assessment, and inclusion in the systematic review. 3.2. Clinical Findings The 23 studies included in the systematic review were published between 2009 and 2024 and were grouped according to exercise modality into endurance training (n= 11), HIIT (n= 7), and acute physiological test studies (n= 5). Collectively, these investigations enrolled a total of 859 participants, predominantly healthy athletes or physically active individuals, with mean ages ranging from 23 to 35 years. The studies were conducted across Europe, Asia, and Oceania, with the highest representation from the United Kingdom, France, Australia, and Germany. Sample sizes ranged from 14 to 50 participants per study, and both male and female athletes were included, though males predominated (approximately 75%). Across all groups, STE—most frequently using General Electric imaging systems—was employed to evaluate LV deformation, torsional mechanics, and diastolic indices in response to different exercise modalities. The eleven endurance studies [20–30] predominantly investigated the acute effects of single prolonged endurance events (such as marathon, ultramarathon, triathlon, or time- trial cycling) rather than long-term
were included, though males predominated (approximately 75%). Across all groups, STE—most frequently using General Electric imaging systems—was employed to evaluate LV deformation, torsional mechanics, and diastolic indices in response to different exercise modalities. The eleven endurance studies [20–30] predominantly investigated the acute effects of single prolonged endurance events (such as marathon, ultramarathon, triathlon, or time- trial cycling) rather than long-term training adaptations. These acute endurance exposures
J. Clin. Med.2025,14, 8210 6 of 26 consistently demonstrated significant post-race reductions in LV strain indices, particularly GLS and GCS (p< 0.05 in [21–23,25–27,29]) (Table). Table 1.Summary of studies [20–30] investigating the acute and subacute effects of endurance exercise on biventricular mechanics using two-dimensional speckle-tracking echocardiography. ACS, apical circumferential strain; cTnT, cardiac troponin T; CRIT, endurance cycling race intervention; GCS, global circumferential strain; GE, General Electric; GLS, global longitudinal strain; GRS, global radial strain; LASr, left atrial reservoir strain; LV, left ventricle; NS, not specified; RASr, right atrial reservoir strain; RV, right ventricle. Study Name, Publication Year and Country Size (%Males) Mean Age (Yrs) Repeated Measures Assessment Software Endurance Exercise Influence on Biventricular Mechanics Nottin S. (2009) France [20] 23 (100) 40 within 3 days before and within 45 min of the race completion GE 14 h triathlon race Decreased LV longitudinal, circumferential and radial strains; slightly reduced/delayed twist; depressed/delayed untwisting George K. (2009) U.K. [21] 19 (100) 41 24 h prior to the race and within 60 min of race finish GE (>40 and <90 frames per second) 89-km Comrades Marathon Post–race significant reduction in LV–GLS, LV–GCS and LV–GRS Chan-Dewar F. (2010) U.K. [22] 14 (100) 32 24 h prior to the race and within 60 min of race completion GE (between 40 and 80 frames per second) 42.2-km London marathon Mild reduction in LV–GLS, significant reduction in ACS; not altered rotation and unchanged LV torsion Oxborough D.L. (2011) U.K. [23] 16 (75) 42 24 h before starting the race, and within 1 h of race completion GE (<90 frames/second) 161-km ultramarathon Reduced LV strain in all planes (longitudinal, circumferential and radial); unchanged LV torsion; LASr impairment; RV dilatation with reduced RV strain. Unnithan V.B. (2015) U.K. [24] 20 (100) 15.2 Prior to and 45–min post–race GE 5 km cross–country race Minor transient decrease in LV–GLS 45–min post–race. Stewart G.M. (2015) Australia [25] 15 (100) 28 1.5 h before and after CRIT60 GE (NS) 60-min endurance cycling intervention (CRIT60) Decreased LV–GLS and RV–GLS; unchanged LV torsion; increased cTnT. Stewart G.M. (2017) Australia [26] 23 (100) 28 Before and after
(100) 15.2 Prior to and 45–min post–race GE 5 km cross–country race Minor transient decrease in LV–GLS 45–min post–race. Stewart G.M. (2015) Australia [25] 15 (100) 28 1.5 h before and after CRIT60 GE (NS) 60-min endurance cycling intervention (CRIT60) Decreased LV–GLS and RV–GLS; unchanged LV torsion; increased cTnT. Stewart G.M. (2017) Australia [26] 23 (100) 28 Before and after 90-min CRIT GE (frame rate of 50–80 frames/sec) 90-min endurance cycling intervention Transient reductions in LV–GLS, LV–GCS and RV–GLS Sengupta S.P. (2018) India [27] 50 (88) 40.8 Before training and within 10 days of completion of marathon GE (frame rate of 50–80 frames/sec) 42.2-km marathon Decreased LV–GLS and LV–GCS; unchanged LV–GRS; increased NT–proBNP Oxborough D.L. (2019) U.K. [28] 23 (100) 27.4 Baseline and after 24–wk training GE (NS) 24-wk endurance training program Unchanged LV–GLS; mild reduction in LV–GCS and LV–GRS; increased basal rotation Pagourelias E.D. (2022) Greece [29] 27 (70.4) 45 24 h before starting the race, and within 10 min after finishing GE (NS) 246 km ultra–marathon running race Decreased LV–GLS and RV–GLS; unchanged LASr and RASr Birat A. (2023) France [30] 20 (100) 15.8 Immediately before and 24-h after the race GE (NS) 68.5-km competitive adventure race Unchanged LV–GLS; reduced LV global myocardial work, LV twisting and untwisting
Description
This systematic review evaluates the effects of various exercise modalities on left ventricular mechanics.