Epicardial Obesity and Myocardial Remodeling After Mitral Valve Surgery

1. Introduction

The process of myocardial remodeling is dynamic and occurs at various levels—ranging from the cellular (cardiomyocytes) to the organ level (alterations in cardiac geometry)—involving changes in the size as well as the structural and functional characteristics of the heart chambers [1,2]. Myocardial remodeling processes in patients with acquired valvular heart disease (VHD) clearly depend on the type and etiology of the defect, the severity of intracardiac and pulmonary circulation hemodynamics, as well as the nature and clinical significance of comorbid conditions [2]. Obesity is currently one of the most prevalent comorbid conditions among individuals with cardiovascular diseases. According to WHO data, the prevalence of obesity has doubled over the past twenty years, and as of 2022, the number of obese individuals worldwide was approaching one billion [3].

A number of studies have demonstrated the contribution of obesity to the process of myocardial remodeling in individuals with cardiovascular diseases [4,5]. However, there is an 'obesity paradox', demonstrating a lower risk of mortality following cardiac surgery in obese individuals compared to those without obesity [6]; this implies the constrained clinical and diagnostic use of body mass index (BMI) as a measure of obesity.

It is important to note that it is the visceral adipose tissue (VAT) that serves as a specific predictor of pathological myocardial remodeling, whereas there is no established correlation between BMI values and the pattern of remodeling [7,8,9].

Researchers are currently showing particular interest in epicardial adipose tissue (EAT) as a component of visceral obesity, due to its immediate proximity to the myocardium and their shared blood supply network.

To date, accumulated data regarding the influence of VAT on the structural and functional characteristics of the heart chambers have been obtained primarily from patients with coronary pathology and heart failure [10]. The influence of visceral obesity components on the cardiac valvular apparatus has been evaluated primarily in experimental studies [11,12]. This determines the relevance of further clinical studies aimed at investigating various components of obesity and their influence on the pattern of myocardial remodeling in patients with acquired VHD [13,14].

Objective of the study: To evaluate the association between epicardial obesity and the pattern of perioperative myocardial remodeling in patients undergoing surgical correction of mitral valve disease.

2. Materials and Methods

A total of 148 patients undergoing surgical correction of non-infectious mitral valve (MV) disease under cardiopulmonary bypass (CPB) were screened for a prospective, observational, non-randomized, single-center study conducted between 2024 and 2025.

All patients provided informed consent to participate in the study. The study was conducted in accordance with the Good Clinical Practice standards and the principles of the Declaration of Helsinki. The study protocol was approved by the Institutional Review Board of the Research Institute for Complex Issues of Cardiovascular Diseases (Minutes No. 7, dated September 16, 2024);

Inclusion criteria: age between 35 and 75 years; presence of acquired mitral valve disease caused by rheumatic heart disease, connective tissue dysplasia syndrome, or degenerative lesions; planned surgical correction of the mitral valve, either isolated or in combination with aortic or tricuspid valve repair, under CPB; signed informed consent to participate in the study.

Exclusion criteria: mitral valve disease etiology associated with infectious endocarditis; hemodynamically significant stenosis of coronary or peripheral arteries requiring revascularization; body weight exceeding 120 kg; tachyform of atrial fibrillation (AF) or atrial flutter (AFL) at the time of examination, preventing multispiral computed tomography (CT) imaging; presence of an implanted pacemaker; absolute contraindications for CT; severe respiratory failure (Grade II–III); pregnancy or lactation; severe liver dysfunction (>9 points on the Child-Pugh scale, Class C); requirement for hemodialysis. history of myocardial infarction, acute cerebrovascular accident, or pulmonary embolism within six months prior to hospitalization for surgical correction; severe neurological deficit; psychiatric disorders.

Baseline characteristics of the patients prior to MV correction are presented in Table 1.

Two-dimensional and three-dimensional transthoracic echocardiography (Echo) was performed prior to the surgery, on days 7 and 30, and one year after MV correction. In addition to the standard echocardiographic protocol, myocardial strain and right ventricular (RV) systolic function were assessed using an expert-class system (Vivid E9, GE Healthcare, USA).

To identify epicardial obesity and quantify EAT volume, patients underwent cardiac CT prior to surgical MV correction. Imaging was performed on a 64-slice scanner (Siemens Somatom, Germany) with with synchronization of ECG recording (Figure 1).

Note: Epicardial adipose tissue is highlighted in yellow

Currently, there is no universally recognized criterion for epicardial obesity (EO). Therefore, ROC analysis was required to identify the threshold for epicardial fat volume that determines the presence or absence of EO. The presence of AF/AFL in patients with mitral valve disease was selected as the factor for determining the EAT threshold via ROC analysis. The choice of AF/AFL as the rhythm disorder was based on the previously confirmed association between epicardial fat and the presence of AF/AFL in patients with cardiovascular diseases [15,16].

In our study, an EAT volume exceeding 115.1 cm^³ (by the MSCT data) was associated with the presence of AF/AFL (AUC = 0.714, 95% CI 0.57–0.83, p = 0.003), according to the ROC analysis, with a sensitivity of 52% and a specificity of 84.6%.

Among the included patients (n = 148), there was an equal distribution of females (47,3%) and males (52,7 %) (median age 65.0 [53.0; 67.8] years).

Based on the obtained EAT threshold, patients were divided into two groups: 62 patients with EO (EAT volume > 115.1 cm^³) and 86 patients without EO (EAT volume ≤ 115.1 cm^³) according to cardiac CT data.

The study design is presented in Figure 2.

Statistical data analysis was performed using standard non-parametric methods with MedCalc software. Quantitative variables are presented as median and interquartile range (Me [LQ; UQ]); qualitative variables are expressed as percentages. The Mann–Whitney U test was used to compare quantitative parameters between groups. Differences in qualitative variables were assessed using Pearson’s chi-squared test. Intragroup changes over time were evaluated using the Wilcoxon signed-rank test. Results of logistic regression analysis are presented as odds ratios (OR) and the level of statistical significance (p-values). Linear regression analysis results are reported as B-coefficients and p-values. Receiver Operating Characteristic (ROC) −curves were used to calculate threshold values (p<0.05, Area Under the Curve (AUC)>0.6). The critical level of statistical significance was set at p<0.05.

3. Results

There were no significant differences between patients with and without epicardial obesity regarding sex (31 females (50.0%] in the EO group vs. 39 (45.3%) in the non-EO group, p=0.35). The groups were also comparable in age (65.0 [61.75; 69.0] years for the EO group vs. 58.5 [51.0; 67.0] years for the non-EO group, (p=0.11)).

CT data indicates that rheumatic etiology of MV disease was more common in patients with EAT volume >115.1 cm^³ (62.9% vs. 34.9%, p=0.04). Connective tissue dysplasia predominated in the non-EO group (51.16% vs. 19.35%, p=0.03).

Patients with EO were twofold less likely to present with MV regurgitation (33.87% vs. 69.8%, p=0.017), while combined MV disease was observed twice as frequently as in patients without EO (41.9% vs. 16.1%, p=0.005). Patients in the EO group had a more severe functional status; notably, while no patients in the non-EO group had NYHA Class IV chronic heart failure (CHF), it was observed in 9.7% of those with EO (p=0.037).

Patients with EO had a higher prevalence of AF/AFL compared to the non-EO group (71.9% vs. 38.4%), p=0.006), and a higher incidence of high-grade ventricular extrasystole (Lown grades IVa–V): 27.4% vs 9.3%, p=0.04. Body mass index was higher in the EO group (32.4 [28.0; 37.1] vs. 23.8 [21.98; 26.6] kg/m^², p<0.001). No other differences in comorbidities were identified.

Preoperatively, patients received standard therapy according to guidelines for chronic heart failure [17]. Significant differences regarding EO were found in the use of beta-blockers and anticoagulants, which were taken by significantly more patients with EAT volume >115.1 cm^³. This finding is likely attributed to the higher prevalence of high-grade ventricular extrasystole and AF/AFL in this group.

The next stage involved assessing the association between EAT volume and the pattern of left and right heart chamber remodeling prior to MV correction. According to Echo results, patients with EO were characterized by larger atrial volumes and a more pronounced decrease in RV systolic function (reduced tricuspid annular plane systolic excursion - TAPSE) (Figure 3).

According to the results of univariate logistic regression analysis, an EAT volume exceeding 115.1 cm^³ is associated with an increased risk of reduced RV systolic function (TAPSE < 1.7 cm) prior to MV correction (OR 4.6 [95% CI 1.2543–16.7481], p=0.021).

In the multivariate analysis conducted to assess the direct contribution of EAT to impaired RV function (TAPSE), the significance of an EAT volume >115.1 cm³ as a predictor of RV dysfunction remained robust (Table 2), even after adjusting for other factors potentially affecting RV systolic function.

However, when AF was included in the analysis, the significance of EAT volume was eliminated, the statistical significance of EAT volume was attenuated. The presence of preoperative AF/AFL in patients with MV disease is associated with a reduction in TAPSE, with a B-coefficient of -0.396 (p < 0.001).

According to the logistic regression analysis, an EAT volume exceeding 93.9 cm^³ (representing the median value for the entire patient cohort) was associated with an increased likelihood of impaired LV GLS (OR 3.75; 95% CI 1.076–13.073, p = 0.038).

When patients were devided into groups by preoperative EAT volume measured by cardiac MSCT, less favorable reverse remodeling was observed during the postoperative follow-up in the group with an EAT volume > 115.1 cm^³ compared to the non-EO group (Table 3). One year after surgery, patients with EO had significantly larger left atrial (LA) diameter, indexed LA volume, and LA area compared to those without EO. During the postoperative period, left ventricular (LV) volumes and LV mass were also significantly higher in patients with EO. In the early postoperative period, patients with EO were characterized by more severely impaired LV global longitudinal strain (GLS) (Table 3).

Regression analysis revealed that 1 cm^³ increase in EAT volume was associated with an increase in right atrial (RA) volume, with a B-coefficient of 0.2759 (p=0.005) at one week and a B-coefficient of 0.2251 (p=0.0303) at one month postoperatively. Furthermore, EAT volume was associated with increased RV end-systolic and end-diastolic volumes as measured by 3D Echo, with B-coefficients of 0.1635 (p=0.016) and 0.2332 (p=0.016), respectively, at one week after the surgery.

Furthermore, each 1 cm^³ increase in EAT volume was associated with an increase in RA area, with a B-coefficient of 0.04221 (p=0.0308), and an increase in RV basal diameter, with a B-coefficient of 0.00321 (p=0.039), at the one-year follow-up after mitral valve correction. One year postoperatively, the proportion of patients with Grade III tricuspid regurgitation was significantly higher in the EO group (p=0.0424) (Table 3).

Postoperative cardiac remodeling dynamics were evaluated in patients following mitral valve correction, stratified by the presence of epicardial obesity (EO).

It was found that, compared to preoperative values, more pronounced reverse remodeling of the left heart chambers was observed in the non-EO group one year postoperatively:

— left ventricular mass index (LVMI) significantly decreased by 28% (p=0.003);

— left ventricular end-diastolic diameter (LVEDD) decreased by 10.3% (p=0.01);

— LA diameter decreased by 9.4% (p=0.002) and left atrial volume index (LAVI) decreased by 33.6% (p=0.004).

In contrast, these parameters showed no significant changes in patients with EO.

Patients without excess epicardial fat demonstrated an improvement in LV GLS of 12.2% (p=0.004) between the one-month and one-year follow-up, while no significant dynamics were observed in the EO group.

The non-EO group was characterized by a 14% decrease in LV end-systolic volume (p=0.04) one month after valve correction, whereas this parameter increased in the EO group (p=0.009). A 15% increase in LV stroke volume (p=0.009) was noted in the non-EO group from the one-month to one-year follow-up.

When evaluating postoperative remodeling dynamics of the right heart chambers, the group with an EAT volume less than 115.1 cm³ showed an increase in RV stroke volume by 76% (p=0.003), RV ejection fraction by 6.5% (p=0.04), and RV fractional area change by 12.2% (p=0.005) from one month to one year, while no significant dynamics were observed in the EO group. In the non-EO group, TAPSE increased from 1.4 [1.2; 1.72] cm to 1.7 [1.5; 1.9] cm (p=0.002) between the 1-month and 1-year postoperative periods, whereas no significant increase was observed in the EO group, which had lower baseline TAPSE values.

When analyzing the incidence of early postoperative complications, a higher frequency of stroke/ transient ischemic attack was observed in the EO group (n=6; 9.7%) during the early postoperative period (p=0.047). This finding may be associated with the significantly higher prevalence of AF/AFL in this group (p=0.014).

One patient developed ventricular fibrillation, which was fatal (6 months after MV surgery). This patient had evidence of EO before MV correction. The death of the second patient was not related to cardiovascular causes.

The incidence of readmissions for HF decompensation and cardiac arrhythmias was significantly higher in the EO group, which may be attributed to a higher prevalence of pre-existing AF in this patient cohort (Table 4).

One year after MV correction, the incidence of arrhythmias (AF/AFL) was significantly higher in the group with EAT volume >115.1 cm^³ compared to the group with lower EAT volume: 29 (48.33%) vs 8 (9.3%) patients, respectively (p=0.003). Among patients discharged from the hospital with AF/AFL (n=54), sinus rhythm restoration during the one-year follow-up period occurred in 12 (22.2%) patients in the EAT group and 19 (22.1%) patients in the non-EAT group (p=0.1625). No significant differences were observed in NYHA functional class between the EAT and non-EAT groups one year after surgery (Table 4).

4. Discussion

Obesity is a well-established driver of myocardial remodeling in patients with cardiovascular diseases, as evidenced by extensive fundamental and applied studies [4,8]. Pathological expansion of VAT and EAT leads to adipocyte hypertrophy and hyperplasia. This triggers an adipokine imbalance and a shift toward a pro-inflammatory profile, resulting in tissue damage, myocardial hypertrophy, and fibrosis [18]. Recent studies indicate that excessive visceral adipose tissue serves as a key predictor of pathological left ventricular remodeling, whereas BMI often shows no correlation with remodeling patterns [7,8,9].

Our study focused specifically on EAT due to its anatomical proximity to the myocardium and its potential impact on adverse cardiac chamber remodeling. Adipokines and other bioactive substances secreted by EAT maintain energy homeostasis and provide mechanical protection for the myocardium. However, its pathological expansion is associated with adipokine dysregulation and the active production of pro-inflammatory interleukins, which may affect cardiomyocytes [4,19]. Research indicates that increased EAT volume is associated with unfavorable outcomes in patients with heart failure with preserved LV ejection fraction [20,21]. Currently, most publications studying structural and functional cardiac impairments in individuals with visceral obesity focus primarily on patients with coronary pathology and chronic heart failure [10,22].

In our study, myocardial remodeling was assessed using both standard transthoracic echocardiography (cardiac chamber dimensions and volumes) and an expanded protocol. The latter included speckle-tracking echocardiography to evaluate ventricular myocardial strain, as well as RV systolic function analysis based on 2D and 3D echocardiographic data. Schulz et al. [21] in their study demonstrated that increased EAT volume was associated with impaired GLS of the left ventricle and left atrium in patients with heart failure and diastolic dysfunction (HFpEF).

Several other studies have also demonstrated a link between EAT and impaired LV systolic and diastolic function, LV enlargement, LA fibrosis, and electrophysiological remodeling, the onset and severity of AF and arterial stiffness parameters in patients with heart failure and ischemic heart disease [23,24,25,26,27]. For the cohort of patients with acquired VHD, available data are limited to associations between EAT thickness and the presence or severity of left-sided heart valve calcification [4], as well as the association between EAT thickness, measured by Echo, and the degree of LV remodeling in moderate ^_ to ^_ severe aortic stenosis [14,28]. Furthermore, in age-associated valvular diseases, EAT thickness is directly correlated with the severity of comorbid conditions, such as diabetes mellitus and obesity, as well as advanced patient age.

There is also evidence, suggesting a correlation between EAT volume (measured by 3D echocardiography) and adverse cardiovascular and neurological events one year after transcatheter aortic valve replacement [29]. Another study demonstrated that patients undergoing transcatheter aortic valve intervention with low EAT volume (<49 cm³ on CT) and high EAT attenuation (≥ -86 HU) had higher all-cause mortality (log-rank p=0.02 and p=0.01, respectively). This association remained significant even after adjusting for age, sex, and clinical characteristics (HR 1.71, p=0.02 and HR 1.73, p=0.03, respectively) [30].

To date, no studies have investigated the relationship between epicardial adipose tissue thickness or volume and the patterns of myocardial remodeling in cardiac surgery patients with MV disease of various etiologies across a wide age range.

The conducted study is the first to demonstrate a significant association between EAT and structural-functional changes in both the left and right cardiac chambers in the cohort of surgical patients with MV disease. This relationship was observed both preoperatively and during the one-year follow-up period following surgical correction.

An EAT volume exceeding 93.9 cm^³, as measured by CT, is associated with a more pronounced impairment of LV global longitudinal strain. Meanwhile, a higher EAT volume (exceeding 115.1 cm^³) is associated with increased atrial volumes and deterioration of the RV systolic function. Notably, the contribution of epicardial obesity to the development of RV dysfunction remains significant even after adjusting for factors such as age, sex, type of valvular heart disease, comorbid profile, and a high probability of pulmonary hypertension based on Echo data.

In the early postoperative period, patients with EO exhibited larger RA dimensions and RVESV, as well as impaired RV-pulmonary artery (RV-PA) coupling. At the one-year follow-up, patients with EO were characterized by larger dimensions of the right-sided cardiac chambers.

Perioperative reverse myocardial remodeling serves as a reflection of the functional status, quality of life, and prognosis of surgical patients with acquired VHD [20]. Our study highlights that, throughout the one-year postoperative period, EAT volume, in particular, exerts a more pronounced contribution to the adverse dynamics of myocardial remodeling. Intragroup dynamics analysis revealed that patients without EO exhibited significant reverse myocardial remodeling across several parameters by the one-year follow-up, along with improvement in RV systolic function. In contrast, these favorable changes were not observed in the group with EO.

5. Conclusions

The presence of epicardial obesity (verified by cardiac CT) in cardiac surgical patients with acquired MV disease is associated with less favorable remodeling patterns of both the left and right cardiac chambers preoperatively, as well as impaired reverse myocardial remodeling throughout the one-year follow-up period. It was demonstrated that the absence of epicardial obesity on cardiac CT in patients undergoing surgical correction is associated with significantly more favorable postoperative reverse LV myocardial remodeling and an improvement in RV systolic function. A timely and comprehensive approach to identifying various manifestations of obesity, combined with a detailed assessment of the remodeling dynamics in both the left and right cardiac chambers, would allow for the modification of factors adversely affecting the disease course in cardiac surgery patients with MV disease at the preoperative stage. The adoption of these advanced preoperative screening algorithms for patients with valvular heart disease could, in turn, enhance treatment efficacy and improve clinical outcomes.