Association of epicardial fat tissue combined with left ventricular strain with left ventricular fibrosis in obese patients: a cardiac magnetic resonance study

Introduction

Obesity is now universally recognized as an independent risk factor for a broad spectrum of cardiovascular diseases (CVDs), encompassing coronary heart disease (CHD), heart failure (HF), and atrial fibrillation (AF) (1-4). Epidemiological data reveal a persistently increasing global prevalence, positioning obesity as a critical public health challenge. Importantly, even individuals with uncomplicated obesity (absence of comorbidities) demonstrate measurable cardiac structural alterations and subclinical functional impairments, which constitute predisposing factors for subsequent HF development (5,6).

Epicardial adipose tissue (EAT), anatomically localized between the visceral pericardium and myocardial surface, exhibits complex physiological roles in metabolic regulation. Under normal conditions, EAT demonstrates functional parallels with brown adipose tissue (BAT), exerting cardioprotective effects through thermogenesis and anti-inflammatory mechanisms. However, in pathological contexts—particularly obesity, metabolic syndrome, and diabetes mellitus—these beneficial properties become dysregulated, with EAT transitioning into a pro-atherogenic entity (7-10). This pathophysiological transformation is characterized by (I) mechanical compression of the myocardium, and (II) promotion of ventricular fibrosis, both of which are strongly associated with CVD progression. Among non-invasive imaging techniques, cardiac magnetic resonance (CMR) is considered the reference standard for quantitative assessment of cardiac structure and function. CMR enables precise volumetric quantification of EAT while simultaneously providing high-resolution evaluation of myocardial deformation. Specifically, late gadolinium enhancement (LGE) imaging within the CMR protocol serves as a crucial diagnostic modality for detecting myocardial fibrosis, particularly in the left ventricular (LV) myocardium, thereby offering valuable insights into fibrotic remodeling associated with EAT dysfunction (11). Recently, Yuan et al. further demonstrated a significant association between CMR-derived EAT and myocardial fibrosis in Duchenne muscular dystrophy patients, providing additional evidence for the role of EAT in fibrotic remodeling (12). CMR feature tracking (CMR-FT) is a post-processing technique that enables quantitative assessment of global or regional LV function using cine sequences without requiring additional dedicated scans. The derived LV strain can predict adverse cardiovascular events in certain diseases.

Early assessment of cardiac structural and functional changes in obese patients is crucial for reducing CVD incidence and improving quality of life. Currently, the association between EAT, LV strain, and LV fibrosis in obese patients remains unclear. This study aimed to comprehensively analyze the relationship between EAT, LV strain, and LV fibrosis in obese patients using CMR techniques. It further investigates the association of combined EAT and LV strain with LV fibrosis, thereby providing imaging evidence for the early clinical diagnosis and intervention of obesity-related LV fibrosis. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0336/rc).

Methods

Study population

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Research and New Technology Ethics Committee of Yijishan Hospital of Wannan Medical College [approval No. 2023 Lun Shen Sheng Yan Di (01)]. Informed consent was provided by all participants. Patients with obesity undergoing CMR examinations at the First Affiliated Hospital of Wannan Medical College (Yijishan Hospital of Wannan Medical College) were prospectively recruited from June 2023 to June 2025. Age- and gender-matched healthy cases with normal body weight were simultaneously enrolled as the control group. The inclusion criteria were as follows: obese patients with a body mass index (BMI) ≥28.0 kg/m²; healthy weight controls with BMI >18.5 and <24.0 kg/m². The exclusion criteria were as follows: (I) history of CVD, such as CHD, valvular heart disease, primary cardiomyopathy, congenital heart disease, or severe arrhythmias; (II) concurrent conditions including diabetes, sleep apnea syndrome, hyperlipidemia, hyperinsulinemia, and so on; (III) concurrent diseases potentially affecting the cardiovascular system, such as malignancies, hematologic disorders, thyroid diseases, or significant hepatic/renal diseases; (IV) severe CMR image artifacts precluding measurement. The inclusion and exclusion criteria are summarized in Figure 1.

Figure 1 Flowchart of the study. BMI, body mass index; CMR, cardiac magnetic resonance; LGE, late gadolinium enhancement.

CMR protocol

All participants underwent scanning using a Siemens Vida 3.0T MR system (Siemens, Erlangen, Germany) with an 18-channel phased array body coil. Cine sequences were acquired using balanced steady-state free precession (b-SSFP) with the following sequences: long-axis two-, three-, and four-chamber, and short-axis views (covering the range from the cardiac base to the apex). LGE sequences were acquired using a reverse recovery gradient echo sequence 8 minutes after contrast agent injection.

CMR analysis

Standard cardiac function parameters

Image post-processing was performed using CVI42 software (Circle Cardiovascular Imaging, Calgary, Alberta, Canada, version 6.0.2), wherein cine short-axis sequences were imported for automatic delineation of end-diastolic and end-systolic endocardial and epicardial contours of the left ventricle and right ventricle, with manual corrections applied as needed. All measured parameters were normalized to body surface area. The derived LV functional parameters included end-diastolic volume, end-systolic volume, stroke volume, ejection fraction, cardiac output, cardiac index, and LV mass. Right ventricular functional analysis included end-diastolic volume, end-systolic volume, stroke volume, ejection fraction, cardiac output, and cardiac index. Left atrial (LA) assessment was performed from two- and four-chamber views, providing minimum volume, maximum volume, and ejection fraction.

Myocardial strain parameters

LV and LA strain parameters were analyzed offline using dedicated CVI42 software. For LV analysis, end-diastolic endocardial and epicardial contours were manually traced on short-axis cine stacks from the base to the apex. Slices where the LV outflow tract was visible were excluded. Global and regional (basal, mid, apical) circumferential and radial strain parameters were automatically derived by the software, including peak circumferential strain and peak radial strain, along with their respective peak systolic and diastolic strain rates. Similarly, longitudinal strain parameters were obtained by contouring the endocardium and epicardium on two-, three-, and four-chamber cine images, yielding peak longitudinal strain (PLS), and its corresponding peak systolic and diastolic strain rates. For LA analysis, end-diastolic endocardial and epicardial contours were manually delineated on two- and four-chamber cine views, with careful exclusion of the pulmonary veins and LA appendage. The software then generated time-strain and strain-rate curves using pixel tracking. Key LA reservoir, conduit, and contractile functions were assessed using the following parameters: reservoir strain, conduit strain, and contractile strain, as well as their corresponding positive peak strain rate, early diastolic strain rate, and late diastolic strain rate.

Assessment of EAT

On cine short-axis images at end-diastole, manual outlining is performed layer by layer. The epicardial visceral layer is traced with red lines, whereas the fat wall layer is outlined with green lines. By setting specific signal intensity thresholds, high-signal fat tissue is marked in yellow to distinguish EAT from surrounding tissues. During the outlining process, particular attention must be paid to exclude coronary artery pathways and pericardial fat from the region of interest (ROI). After the delineation of EAT at all layers is completed, the system automatically calculates and outputs the epicardial adipose tissue volume (EATV). The EAT thickness measurement method is illustrated in Figure 2. At end-diastole in cine long-axis four-chamber view, the EAT thickness is measured at the left atrioventricular groove (LAVG), right atrioventricular groove, and anterior interventricular groove (AIVG). In end-diastole cine short-axis basal images, EAT thickness is measured at the inferior interventricular groove (IIVG), superior interventricular groove (SIVG), and the right ventricular free wall, with the average value of three different measurement points taken for the thickness of the right ventricular free wall. During the actual measurement process, some participants could not complete simultaneous measurements of the three thicknesses due to variations in cardiac anatomy or limitations in image quality, which resulted in poor visualization of the basal segment. In such cases, an alternative measurement level can be selected—specifically, a basal-level short-axis cine image that is anatomically closest and has the clearest imaging quality.

Figure 2 Acquisition of epicardial fat thickness, volume, and myocardial strain. Measure end-diastolic EAT thickness at the horizontal long-axis four-chamber view for the RAVG, LAVG and AIVG (A). Measure EAT thickness at the short-axis view for the IIVG, SIVG and RVFW (B), with RVFW averaged from three measurement points. Manually measure the EATV, located between the red and green curves (C). PCS (D); PLS (E); PRS (F). AIVG, anterior interventricular groove; EAT, epicardial adipose tissue; EATV, epicardial adipose tissue volume; IIVG, inferior interventricular groove; LAVG, left atrioventricular groove; PCS, peak circumferential strain; PLS, peak longitudinal strain; PRS, peak radial strain; RAVG, right atrioventricular groove; RVFW, right ventricular free wall; SIVG, superior interventricular groove.

Assessment of LV fibrosis

Myocardial fibrosis was defined as regions with LGE signal intensity exceeding six standard deviations above that of normal myocardium (13). LV LGE images in patients with obesity were independently assessed by two associate chief radiologists (Physician 1 and Physician 2), each with at least 3 years of cardiac magnetic resonance imaging (MRI) experience and both blinded to clinical data. Disagreements were resolved by consensus with a third senior radiologist (8 years of experience). For reproducibility analysis, 20 randomly selected cases (10 patients and 10 controls) were re-evaluated by Physician 1 after 30 days using the same method to determine intra-observer agreement (Table S1).

Statistical analysis

Statistical analyses were conducted using the software SPSS 27.0 (IBM Corp., Armonk, NY, USA) and R 4.3.2 (R Foundation for Statistical Computing, Vienna, Austria). Normality tests for quantitative data: normal distribution (x¯±s) with independent samples t-test; non-normal distribution median (Q1, Q3) with Mann-Whitney U test. Kruskal-Wallis H was employed test for intergroup comparisons, with Bonferroni correction for pairwise comparisons. Categorical data by counts, Chi-squared test for intergroup comparisons. Spearman’s correlation analysis for epicardial fat parameters and myocardial strain parameters, results as correlation matrix heatmap. The association between epicardial fat parameters, LV strain, and LV fibrosis was analyzed using logistic regression. We performed sensitivity analyses to assess the robustness of the core findings (EATV, LAVG, and PLS) against potential confounding by BMI, age, C-reactive protein (CRP), and sex. Each confounder was added individually to a reduced model containing only EATV, LAVG, and PLS. Finally, the discriminative ability of epicardial fat parameters combined with LV strain for LV fibrosis was evaluated using receiver operating characteristic (ROC) curves. Intraclass correlation coefficient (ICC) was used for consistency analysis (ICC >0.75 indicating good consistency). A P value <0.05 was considered statistically significant.

Results

Patient population

This study ultimately enrolled 80 obese patients (28 males) and 80 healthy cases with normal body weight. Patients with obesity were further divided into two groups based on the presence of LV fibrosis: the group with LV fibrosis [LGE(+), n=33] and the group without LV fibrosis [LGE(−), n=47]. Table 1 shows that after stratifying the study population by obesity status, baseline characteristics revealed that most serum markers were higher in obese patients than in healthy cases, with particularly significant differences in myocardial injury markers and inflammatory indicators.

Comparison of conventional cardiac function

As shown in Table 2, the end-diastolic volume, end-systolic volume, and stroke volume of both ventricles in patients with obesity were greater than those in the control group, whereas the cardiac index and ejection fraction of both ventricles were lower than those in the control group, and the differences were statistically significant (P<0.05). The LV cardiac output and LV mass of patients with obesity were increased compared to the control group, but there was no significant difference in right ventricular cardiac output compared to the control group. In addition, the LA minimum and maximum volumes of patients with obesity were greater than those of the control group, but the LA ejection fraction was lower than that of the control group.

Comparison of myocardial strain parameters

As shown in Table 3, regarding LA strain, both strain and strain rate were significantly reduced in obese patients compared to the control group. In LV strain analysis, globally reduced strain and strain rate were observed across all three directional components in the obese group. Segmental analysis confirmed these impairments in the middle and apical segments in all directions. At the basal segment, however, PLSR-S and PLSR-D were preserved, with all other parameters being significantly lower than in controls.

Comparison of subgroups based on epicardial fat parameters and LV strain

As shown in Figure 3, regarding epicardial fat parameters: EAT thickness at all measured locations (RAVG and LAVG, AIVG, IIVG, and SIVG, and right ventricular free wall) and EATV in the LGE(+) group were all significantly higher than those in the control group, with statistically significant differences. Compared with the LGE(−) group, the LGE(+) group exhibited statistically significant increases in LAVG thickness, IIVG thickness, and EATV, whereas no significant differences were observed at other measured locations. Regarding LV strain: Compared with the LGE(−) group, the LGE(+) group demonstrated statistically significant reductions in radial strain, circumferential strain, and longitudinal strain.

Figure 3 Comparison of subgroups based on epicardial fat parameters and left ventricular strain. Green bars represent the control group; orange bars represent the LGE(−) group; red bars represent the LGE(+) group; Asterisks above bars denote statistical significance between connected groups, with * indicating P<0.05. Groups without connecting lines show no statistically significant difference. AIVG, anterior interventricular groove; EATV, epicardial adipose tissue volume; IIVG, inferior interventricular groove; LAVG, left atrioventricular groove; LGE, late gadolinium enhancement; PCS, peak circumferential strain; PLS, peak longitudinal strain; PRS, peak radial strain; RAVG, right atrioventricular groove; RVFW, right ventricular free wall; SIVG, superior interventricular groove.

Correlation analysis between epicardial fat and LV myocardial strain

The correlation between epicardial fat and LA myocardial strain parameters was explored, as shown in Figure 4: EATV and LAVG thickness showed negative correlations with LA reservoir strain, conduit strain, and radial strain (r=−0.34 to −0.26, P<0.05), while exhibiting positive correlations with longitudinal strain and circumferential strain (r=0.18 to 0.36, P<0.05). Longitudinal strain showed positive correlations with right atrioventricular groove, AIVG, IIVG, and SIVG thicknesses (r=0.21 to 0.31, P<0.05).

Figure 4 Correlation analysis. The differently colored squares in the figure illustrate the correlation between the two variables, where green indicates a positive correlation, pink indicates a negative correlation, and the numerical values represent the correlation coefficients. The color scale indicates the correlation coefficient (r), with green representing positive correlations and pink representing negative correlations. *, P<0.05; ***, P<0.005; ****, P<0.001. AIVG, anterior interventricular groove; Ea, booster pump strain; EATV, epicardial adipose tissue volume; Ee, conduit strain; Es, reservoir strain; IIVG, inferior interventricular groove; LAVG, left atrioventricular groove; PCS, peak circumferential strain (left ventricular strain—global); PLS, peak longitudinal strain (left ventricular strain—global); PRS, peak radial strain (left ventricular strain—global); RAVG, right atrioventricular groove; RVFW, right ventricular free wall; SIVG, superior interventricular groove.

Relationship between EAT and LV fibrosis

As shown in Table 4, univariate logistic regression analysis with EAT thickness, EATV, and LV global strain as independent variables and LV fibrosis as the dependent variable revealed that EATV, LAVG thickness, IIVG thickness, radial strain, circumferential strain, and longitudinal strain were predictive factors for LV fibrosis (P<0.05). Subsequently, using the statistically significant indicators from the univariate logistic regression as independent variables and LV fibrosis as the dependent variable, multivariate logistic regression analysis revealed that EATV, LAVG thickness, and longitudinal strain are independent predictors of LV fibrosis (P<0.05).

To validate the robustness of the core predictive model and further explore the role of systemic obesity indicators, we conducted a sensitivity analysis incorporating BMI into the adjusted logistic regression model alongside the independent predictors EATV, LAVG, and PLS identified in the primary multivariate analysis. As shown in Table S2, after mutual adjustment, EATV [adjusted odds ratio (OR) =1.065; 95% confidence interval (CI): 1.031–1.101; P<0.001], LAVG (adjusted OR =1.399; 95% CI: 1.063–1.841; P=0.016), and PLS (adjusted OR =1.420; 95% CI: 1.162–1.735; P<0.001) retained their significant independent predictive value for LV fibrosis, and BMI also demonstrated independent predictive utility (adjusted OR =0.895; 95% CI: 0.814–0.984; P=0.023). Sensitivity analyses adjusting for age, CRP, and sex also yielded consistent results (Tables S3-S5). This indicates that after controlling for localized fat and myocardial function factors within a multivariate framework, systemic obesity level remains independently associated with myocardial fibrosis.

As demonstrated by the above analysis, EATV, LAVG thickness, and longitudinal strain are independent predictors of LV fibrosis. As shown in Figure 5, ROC analysis revealed that the area under the curve (AUC) for EATV, LAVG thickness, longitudinal strain, and the combined model was 0.827, 0.775, 0.849, and 0.943, respectively. The corresponding sensitivities were 0.758, 0.848, 0.788, and 0.788; specificities were 0.891, 0.674, 0.826, and 0.957; Youden indices were 0.649, 0.522, 0.614, and 0.744; and optimal cutoffs were 121.74 (EATV, mL), 10.90 (LAVG thickness, mm), −8.89 (longitudinal strain, %), and 0.65 (combined model). In this exploratory analysis, the combined model showed the highest AUC of 0.943.

Figure 5 ROC curve for predicting LV fibrosis. AUC, area under the curve; EATV, epicardial adipose tissue volume; LAVG, left atrioventricular groove; LV, left ventricular; PLS, peak longitudinal strain.

Discussion

This study comprehensively evaluated the relationship between EAT and LV fibrosis in patients with simple obesity, as well as associated myocardial microstructural alterations. Compared with controls, the obesity group showed significant impairments in both conventional cardiac function and myocardial strain parameters. EAT volume and regional thickness were also significantly increased in these patients. When stratified by the presence of LV fibrosis, the LGE(+) group exhibited significantly higher EAT thickness at all measured locations and EATV compared to the LGE(−) group and the control group. Compared with the LGE(−) group, the LGE(+) group exhibited statistically significant increases in LAVG thickness, IIVG thickness, and EATV, and decreases in radial strain, circumferential strain, and longitudinal strain. EATV, LAVG thickness, and longitudinal strain were independently associated with LV fibrosis and showed promising discriminative ability in this exploratory analysis. A recent comprehensive review by Matusik et al. highlights the contemporary role of CMR in myocardial tissue characterization, fibrosis assessment, and cardiovascular risk stratification (14). Our findings on EAT, LV strain, and LGE-based fibrosis align with this broader CMR framework, further supporting the use of CMR for early detection of obesity-related cardiac remodeling.

Myocardial strain provides a more sensitive reflection of early ventricular function changes compared to LVEF (15,16). Even in obese individuals without diagnosed CVD, subclinical abnormalities in myocardial deformation can be detected. Previous studies have indicated that in patients with normal LVEF, both circumferential and longitudinal myocardial strain in the LV are reduced in obese cases compared to healthy controls, suggesting altered systolic myocardial motion in the LV (17-19). Even in metabolically healthy obese individuals, higher BMI is associated with an increased risk of LV diastolic dysfunction (20). These findings resonate with our study results, which revealed that both global and segmental longitudinal, circumferential, and radial strain and strain rate were lower in the LV of obese patients compared to controls. Consequently, increased BMI in obese patients leads to varying degrees of LV dysfunction, potentially elevating the risk of CVD. Based on our analysis of LV strain, we further investigated differences in LA strain and strain rate between the two groups. We found that both LA strain and strain rate were lower in patients with obesity compared to healthy controls, indicating impaired LV function in this population. Therefore, early detection of LV dysfunction in individuals with obesity is crucial for reducing the incidence of CVD. Furthermore, studies have demonstrated that early intervention targeting obesity and achieving BMI reduction can effectively improve cardiac structure and function in severely obese patients, thereby lowering the risk of CVD and exerting a preventive effect (21-24).

EAT is a distinct visceral fat compartment located between the visceral pericardium and the myocardium, directly infiltrating the heart muscle (24,25). It secretes multiple cytokines, and changes in EAT content correlate with increases in BMI. Ultrasound, computed tomography (CT), and CMR can all measure EAT. However, ultrasound accuracy is compromised by factors such as acoustic window effects and respiration, whereas CT involves ionizing radiation, limiting its repeated use. CMR offers advantages such as excellent tissue contrast and absence of ionizing radiation, enabling standardized quantification of EAT. As highlighted in a recent comprehensive review by Baichoo et al., cardiac adipose tissue imaging plays an increasingly important role in the assessment of coronary artery disease and cardiovascular risk stratification (26). Currently, CMR-LGE serves as the gold standard for in vivo identification and assessment of LV myocardial fibrosis. Increased EAT thickness and volume hold value in cardiac metabolic risk assessment and are associated with higher risks of acute coronary syndrome and AF following cardiac surgery. The early detection of myocardial fibrosis in patients with obesity holds significant importance. To further investigate the correlation between EAT and LV fibrosis in these patients, we conducted a subgroup analysis based on the presence or absence of LV fibrosis. The results showed that the LGE(+) group exhibited greater EAT thickness in all directions and higher EATV compared to the LGE(−) group. Among these, differences in LAVG thickness, IIVG thickness, and EATV were statistically significant. We then explored the correlation between EAT and global LV strain. We found that as LAVG thickness and EATV increased, LA reservoir strain, conduit strain, and radial strain gradually decreased in obese patients, whereas longitudinal strain progressively increased. We hypothesize that EAT, being in close contact with the myocardial surface, can directly infiltrate the myocardium (27,28). Excessive EAT may secrete large amounts of pro-inflammatory cytokines, such as tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6). These inflammatory substances can readily diffuse into the adjacent LA myocardial tissue, triggering inflammation. Inflammation directly impairs the function of cardiomyocytes, disrupting their normal contraction and relaxation. Large numbers of inflammatory cells infiltrate the necrotic area to clear necrotic tissue, while fibroblasts migrate to the necrotic area and secrete extracellular matrix components such as collagen, gradually forming fibrous scar tissue, thereby leading to myocardial fibrosis (29). Increased myocardial fibrosis causes the atrial wall to stiffen, impairing its movement and contraction function, which manifests as reduced reservoir strain and conduit strain, ultimately leading to LA remodeling (30). Nevertheless, without T1 mapping or extracellular volume (ECV), our LGE-based assessment likely underestimates the true extent of diffuse myocardial fibrosis in this obese population. It is noteworthy that the inflammatory environment of EAT is not confined to the LA; it can also envelop and affect the LV. Furthermore, obesity itself leads to increased systemic vascular resistance, thereby elevating LV ejection resistance. This elevation is mediated in part by increased sympathetic nervous system activity and circulating catecholamines, as described by Landsberg and Young (31). Over time, with substantial accumulation of EAT and under the influence of LA damage, impaired LV myocardial function similarly results in fibrosis of the LV myocardium, manifested as a decrease in radial strain. Furthermore, due to LA stiffness and impaired function, the LV initiates compensatory mechanisms to overcome inadequate filling and maintain cardiac output (32). One key pathway involves enhancing its ability for active relaxation and expansion during early diastole, manifested as an increase in longitudinal strain. This coexisting state of compensation and decompensation characteristically reflects cardiac progression toward dysfunction under metabolic risk factors (obesity, increased EAT). The positive correlation between LAVG and IIVG thicknesses with longitudinal strain highlights the position-specific effects of EAT. However, based on current data, these observations remain speculative and warrant further investigation.

Myocardial fibrosis has been extensively studied for its role in disease prognosis assessment, with the extent of fibrosis influencing patient treatment (33,34). Furthermore, the early identification of myocardial fibrosis has emerged as a potential therapeutic target for certain clinical conditions, making the ability to predict and address fibrosis at an early stage particularly crucial. However, LGE sequence scans require gadolinium contrast agents, which may cause adverse reactions such as syncope, allergic reactions, and renal injury. This prevents some patients from undergoing LGE scans due to contrast allergies, hindering the accurate assessment of myocardial fibrosis. Furthermore, this examination is invasive and incurs additional costs. Against the backdrop of increasing emphasis on personalized and precision medicine, we validated the potential of combining EAT with LV strain to predict LV fibrosis. Through logistic regression analysis and ROC curve plotting, the results demonstrated that EATV, LAVG thickness, and longitudinal strain are independent predictors of LV fibrosis with good predictive performance (AUCs of 0.827, 0.775, and 0.849, respectively). The combined model significantly enhances predictive performance. EATV, LAVG thickness, and longitudinal strain serve as non-invasive imaging biomarkers for myocardial fibrosis in obese patients, offering an alternative assessment method for those unable to undergo contrast-enhanced scanning. This approach also opens new avenues for disease research, helps reduce healthcare costs and avoid invasive procedures, while providing imaging evidence for early intervention, prognosis assessment, and clinical management.

Furthermore, a pivotal finding of this study is that BMI retained independent predictive value for LV fibrosis after adjustment for epicardial fat parameters and LV strain (P=0.023). This apparent paradox—where BMI was nonsignificant in univariable analysis but emerged as significant in the multivariable model—reflects the complex pathophysiological interplay underlying obesity-related myocardial remodeling. We hypothesize that a “suppressor variable” effect may be operative: as an indicator of systemic obesity, BMI confers potential cardiotoxicity through pathways such as systemic inflammation, insulin resistance, and hemodynamic load; however, this detrimental influence may be partially masked by its strong association with epicardial fat accumulation. In univariable analysis, these opposing effects likely counterbalance each other, resulting in a nonsignificant net association. After adjusting for epicardial fat—a potent local mediator—the independent adverse effect of BMI, disentangled from localized fat burden, was effectively isolated and clearly demonstrated.

This observation carries dual implications. First, it reinforces that epicardial fat and LV strain represent more direct and powerful risk markers than global obesity severity, given that their inclusion substantially improved the model’s discriminative ability in this cohort. Second, it suggests that systemic obesity and localized cardiac fat may promote myocardial fibrosis through partly distinct mechanisms. Thus, in cardiac risk stratification of obese patients, although overall obesity assessed by BMI remains relevant, the use of cardiac MRI to precisely quantify regional fat distribution and myocardial function provides indispensable incremental information, enabling more accurate risk assessment.

Limitations

This study has several limitations. First, as a single-center study, its generalizability may be limited. In addition, the absence of external or internal validation means that our findings should be considered hypothesis-generating rather than definitive for clinical prediction. Second, the small sample size may have reduced statistical power; future research should include larger, multicenter cohorts. Third, myocardial fibrosis was defined solely by LGE, which detects only focal fibrosis. Diffuse myocardial fibrosis, which may be more relevant in obesity-related cardiomyopathy, cannot be assessed without T1 mapping or ECV quantification. The lack of these complementary tissue characterization techniques limits the biological interpretation of our findings. Future studies should combine LGE with T1 mapping and ECV to fully capture the fibrotic burden. Fourth, the grouping strategy may introduce bias due to factors such as LGE positivity variability, scanning timing, contrast agent use, and image quality, which will be optimized in future studies. Fifth, due to the cross-sectional design of this study, a longitudinal analysis is lacking. Future prospective longitudinal studies are needed to confirm the temporal relationship between EAT accumulation, LV strain impairment, and the development of LV fibrosis.

Conclusions

CMR enables precise assessment of EAT and LV strain in obese patients. In this study, LAVG thickness, EATV, and longitudinal strain were independently associated with LV fibrosis and may provide useful imaging information, but external validation is required before clinical application.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0336/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0336/dss

Funding: This work was supported by the Anhui Province Clinical Medicine Research Transformation Project (No. 202304295107020003).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0336/coif). All authors report that this work was supported by the Anhui Province Clinical Medicine Research Transformation Project (No. 202304295107020003). The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Research and New Technology Ethics Committee of Yijishan Hospital of Wannan Medical College [approval No. 2023 Lun Shen Sheng Yan Di (01)]. Informed consent was taken from all the patients.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Trends in adult body-mass index in 200 countries from 1975 to 2014: a pooled analysis of 1698 population-based measurement studies with 19·2 million participants. Lancet 2016;387:1377-96.
2. Afshin A, Forouzanfar MH, Reitsma MB, Sur P, Estep K, et al. Health Effects of Overweight and Obesity in 195 Countries over 25 Years. N Engl J Med 2017;377:13-27.
3. Kivimäki M, Kuosma E, Ferrie JE, Luukkonen R, Nyberg ST, Alfredsson L, et al. Overweight, obesity, and risk of cardiometabolic multimorbidity: pooled analysis of individual-level data for 120 813 adults from 16 cohort studies from the USA and Europe. Lancet Public Health 2017;2:e277-85. [Crossref] [PubMed]
4. Sandhu RK, Ezekowitz J, Andersson U, Alexander JH, Granger CB, Halvorsen S, Hanna M, Hijazi Z, Jansky P, Lopes RD, Wallentin L. The ‘obesity paradox’ in atrial fibrillation: observations from the ARISTOTLE (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation) trial. Eur Heart J 2016;37:2869-78. [Crossref] [PubMed]
5. Ünlü S, Taçoy G. Early adulthood obesity is associated with impaired left ventricular and right ventricular functions evaluated by speckle tracking and 3D echocardiography. Turk Kardiyol Dern Ars 2021;49:312-20. [Crossref] [PubMed]
6. Golinska-Grzybala K, Wiechec M, Golinski B, Rostoff P, Szlósarczyk B, Gackowski A, Nessler J, Konduracka E. Subclinical cardiac performance in obese and overweight women as a potential risk factor of preeclampsia. Pregnancy Hypertens 2021;23:131-5. [Crossref] [PubMed]
7. Janssen-Telders C, Eringa EC, de Groot JR, de Man FS, Handoko ML. The role of epicardial adipose tissue remodelling in heart failure with preserved ejection fraction. Cardiovasc Res 2025;121:860-70. [Crossref] [PubMed]
8. Konwerski M, Gąsecka A, Opolski G, Grabowski M, Mazurek T. Role of Epicardial Adipose Tissue in Cardiovascular Diseases: A Review. Biology (Basel) 2022;11:355. [Crossref] [PubMed]
9. Rabkin SW. Epicardial fat: properties, function and relationship to obesity. Obes Rev 2007;8:253-61. [Crossref] [PubMed]
10. Chechi K, Richard D. Thermogenic potential and physiological relevance of human epicardial adipose tissue. Int J Obes Suppl 2015;5:S28-34. [Crossref] [PubMed]
11. Wang X, Pu J. Recent Advances in Cardiac Magnetic Resonance for Imaging of Acute Myocardial Infarction. Small Methods 2024;8:e2301170. [Crossref] [PubMed]
12. Yuan W, Xu H, Yu L, Wen L, Xu K, Xie L, Xu R, Fu H, Liu B, Xu T, Zhou X, Bi X, Cai X, Guo Y. Association of increased epicardial adipose tissue derived from cardiac magnetic resonance imaging with myocardial fibrosis in Duchenne muscular dystrophy: a clinical prediction model development and validation study in 283 participants. Quant Imaging Med Surg 2024;14:736-48. [Crossref] [PubMed]
13. Xu J, Yang W, Zhao S, Lu M. State-of-the-art myocardial strain by CMR feature tracking: clinical applications and future perspectives. Eur Radiol 2022;32:5424-35. [Crossref] [PubMed]
14. Matusik PS, Mikrut K, Bryll A, Popiela TJ, Matusik PT. Cardiac Magnetic Resonance Imaging in Diagnostics and Cardiovascular Risk Assessment. Diagnostics (Basel) 2025;15:178. [Crossref] [PubMed]
15. Bolz C, Blaszczyk E, Mayr T, Lim C, Haufe S, Jordan J, Barckow P, Gröschel J, Schulz-Menger J. Adiposity influences on myocardial deformation: a cardiovascular magnetic resonance feature tracking study in people with overweight to obesity without established cardiovascular disease. Int J Cardiovasc Imaging 2024;40:643-54. [Crossref] [PubMed]
16. Homsi R, Yuecel S, Schlesinger-Irsch U, Meier-Schroers M, Kuetting D, Luetkens J, Sprinkart A, Schild HH, Thomas DK. Epicardial fat, left ventricular strain, and T1-relaxation times in obese individuals with a normal ejection fraction. Acta Radiol 2019;60:1251-7. [Crossref] [PubMed]
17. Zhu L, Gu S, Wang Q, Zhou X, Wang S, Fu C, Yang W, Wetzl J, Yan F. Left ventricular myocardial deformation: a study on diastolic function in the Chinese male population and its relationship with fat distribution. Quant Imaging Med Surg 2020;10:634-45. [Crossref] [PubMed]
18. Shen MT, Guo YK, Liu X, Ren Y, Jiang L, Xie LJ, Gao Y, Zhang Y, Deng MY, Li Y, Yang ZG. Impact of BMI on Left Atrial Strain and Abnormal Atrioventricular Interaction in Patients With Type 2 Diabetes Mellitus: A Cardiac Magnetic Resonance Feature Tracking Study. J Magn Reson Imaging 2022;55:1461-75. [Crossref] [PubMed]
19. Rozenbaum Z, Topilsky Y, Khoury S, Pereg D, Laufer-Perl M. Association of body mass index and diastolic function in metabolically healthy obese with preserved ejection fraction. Int J Cardiol 2019;277:147-52. [Crossref] [PubMed]
20. Lewis AJM, Abdesselam I, Rayner JJ, Byrne J, Borlaug BA, Neubauer S, Rider OJ. Adverse right ventricular remodelling, function, and stress responses in obesity: insights from cardiovascular magnetic resonance. Eur Heart J Cardiovasc Imaging 2022;23:1383-90. [Crossref] [PubMed]
21. Ruano-Campos A, Cruz-Utrilla A, López-Antoñanzas L, Luaces M, Pérez de Isla L, Rubio Herrera MÁ, Torres García A, Sánchez-Pernaute A. Evaluation of Myocardial Function Following SADI-S. Obes Surg 2021;31:3109-15. [Crossref] [PubMed]
22. Piché ME, Clavel MA, Auclair A, Rodríguez-Flores M, O’Connor K, Garceau P, Rakowski H, Poirier P. Early benefits of bariatric surgery on subclinical cardiac function: Contribution of visceral fat mobilization. Metabolism 2021;119:154773. [Crossref] [PubMed]
23. de Witte D, Wijngaarden LH, van Houten VAA, van den Dorpel MA, Bruning TA, van der Harst E, Klaassen RA, Niezen RA. Improvement of Cardiac Function After Roux-en-Y Gastric Bypass in Morbidly Obese Patients Without Cardiac History Measured by Cardiac MRI. Obes Surg 2020;30:2475-81. [Crossref] [PubMed]
24. Ayton SL, Gulsin GS, McCann GP, Moss AJ. Epicardial adipose tissue in obesity-related cardiac dysfunction. Heart 2022;108:339-44. [Crossref] [PubMed]
25. van Woerden G, van Veldhuisen DJ, Westenbrink BD, de Boer RA, Rienstra M, Gorter TM. Connecting epicardial adipose tissue and heart failure with preserved ejection fraction: mechanisms, management and modern perspectives. Eur J Heart Fail 2022;24:2238-50. [Crossref] [PubMed]
26. Baichoo R, Fu J, Xiong X, Zhou Q, Guan X, Xu X. The role of cardiac adipose tissue in coronary artery disease: a narrative review of cardiac imaging studies. Quant Imaging Med Surg 2024;14:9685-97. [Crossref] [PubMed]
27. Gorter TM, van Woerden G, Rienstra M, Dickinson MG, Hummel YM, Voors AA, Hoendermis ES, van Veldhuisen DJ. Epicardial Adipose Tissue and Invasive Hemodynamics in Heart Failure With Preserved Ejection Fraction. JACC Heart Fail 2020;8:667-76. [Crossref] [PubMed]
28. Chahine Y, Askari-Atapour B, Kwan KT, Anderson CA, Macheret F, Afroze T, Bifulco SF, Cham MD, Ordovas K, Boyle PM, Akoum N. Epicardial adipose tissue is associated with left atrial volume and fibrosis in patients with atrial fibrillation. Front Cardiovasc Med 2022;9:1045730. [Crossref] [PubMed]
29. Seki H, Nakanishi K, Daimon M, Hirose K, Mukai Y, Yoshida Y, Nakao T, Morita H, Di Tullio MR, Homma S, Komuro I. Epicardial fat accumulation and left heart remodelling in patients with chronic coronary syndrome. Eur Heart J Open 2023;3:oeac082. [Crossref] [PubMed]
30. Schulz A, Backhaus SJ, Lange T, Evertz R, Kutty S, Kowallick JT, Hasenfuß G, Schuster A. Impact of epicardial adipose tissue on cardiac function and morphology in patients with diastolic dysfunction. ESC Heart Fail 2024;11:2013-22. [Crossref] [PubMed]
31. Landsberg L, Young JB. The role of the sympathetic nervous system and catecholamines in the regulation of energy metabolism. Am J Clin Nutr 1983;38:1018-24. [Crossref] [PubMed]
32. Lange T, Backhaus SJ, Schulz A, Hashemi D, Evertz R, Kowallick JT, Hasenfuß G, Kelle S, Schuster A. CMR-based cardiac phenotyping in different forms of heart failure. Int J Cardiovasc Imaging 2024;40:1585-96. [Crossref] [PubMed]
33. da Silva LM, Coy-Canguçu A, Paim LR, Bau AA, Nicolela Geraldo Martins C, Pinheiro S, Citeli Ribeiro V, Magalhães Rocha WE, Mattos-Souza JR, Schreiber R, Antunes-Correa L, Sposito A, Nadruz W Jr, Ramos CD, Neilan T, Jerosch-Herold M, Coelho-Filho OR. Impaired Cardiac Sympathetic Activity Is Associated With Myocardial Remodeling and Established Biomarkers of Heart Failure. J Am Heart Assoc 2024;13:e035264. [Crossref] [PubMed]
34. Zhang X, Yang S, Hao S, Li J, Qiu M, Chen H, Huang Y. Myocardial fibrosis and prognosis in heart failure with preserved ejection fraction: a pooled analysis of 12 cohort studies. Eur Radiol 2024;34:1854-62. [Crossref] [PubMed]