Epicardial Fat and Left Ventricular Diastolic Dysfunction
Epicardial Fat and Left Ventricular Diastolic Dysfunction
Of the 149 subjects in our sample population, 65 were diagnosed with LVDD, and the remainder served as controls. The baseline characteristics of the participants in both groups are summarized in Table 1. Consistent with previous reports on hypertension, the subjects with LVDD were predominantly female, older, and suffered more frequently from hypertension or hyperlipidemia (higher LDL level). Patients with LVDD had significantly higher hsCRP levels (Table 1). The cause of renal failure and the residual renal function did not differ between the groups. Relative to the control group, subjects with LVDD had larger end-diastolic and systolic LV volumes (p < 0.05), greater LA diameters, and larger indexed LV mass values (p < 0.05). Comparison of the functional parameters showed a prolonged deceleration time (DT) (p < 0.05), increased mitral inflow late filling wave (p < 0.001), decreased mitral inflow E/A ratio (p < 0.005), and decreased peak annular early diastolic velocity of the lateral mitral annulus in tissue Doppler imaging (p < 0.001) among the patients with LVDD.
Comparison of the anthropometric characteristics showed higher levels of markers reflecting fat distribution, such as the amounts of total, subcutaneous, visceral, and peritoneal fat (p < 0.001, p < 0.001, p < 0.01, and p < 0.005, respectively), in the LVDD group.
The bivariate Pearson correlation coefficients for LV diastolic function parameters and EpF are shown in Figure 2. EpF was significantly associated with tissue Doppler e', E/e', and DT (r = −0.39, p < 0.001; r = 0.27, p = 0.001; r = 0.29, p < 0.001, respectively) (Figure 2A–C). EpF thickness was greater in patients with LVDD (n = 65; 5.1 ± 2.6 mm) than in controls (n = 84; 2.8 ± 1.6 mm, p < 0.001, Figure 2D).
(Enlarge Image)
Figure 2.
Correlation between epicardial fat (EpF) thickness and left ventricular diastolic dysfunction (LVDD). EpF thickness correlated significantly with (A) e' (r = −0.39, p < 0.001), (B) E/e' (r = 0.27, p < 0.001), and (C) mitral inflow deceleration time (r = 0.29, p = 0.002). (D) EpF thickness in patients with or without left ventricular diastolic dysfunction (LVDD). The bar graph shows the mean + standard deviation. EpF thickness was significantly greater in patients with LVDD (5.1 ± 2.6 cm) than in controls (2.8 ± 1.6 cm, p < 0.001).
We performed univariate analysis to determine the risk factors associated with the development of LVDD. Hypertension, the natural logarithm of the LDL level [Ln(LDL)], Ln(Visceral fat), Ln(Peritoneal fat), the hsCRP level, Ln(LV mass index) and EpF thickness were significantly associated with the presence of LVDD (Table 2).
We previously found subclinical inflammation (e.g., the plasma hsCRP level) to be an independent risk factor for LVDD in patients with metabolic syndromes. We therefore performed multivariate analysis of the risk factors other than adiposity that were associated with LVDD in our univariate analyses and found that the hsCRP level remained associated with LVDD after adjustment for age, gender, diabetes mellitus, hypertension, the hsCRP and LDL levels, and the LV mass index (OR: 1.75, 95% CI: 1.08–3.12, p = 0.038; Table 3, model 1).
We further evaluated fat distribution and its association with LVDD by multivariate regression analysis. Of the CT measures of fat distribution, Ln(visceral fat) and Ln(peritoneal fat) remained associated with LVDD after adjustment for the abovementioned confounding factors (adjusted OR = 2.29, 95% CI = 1.06–5.57, p = 0.013 and adjusted OR = 2.45, 95% CI = 1.1–5.9, p = 0.011 respectively; Table 3, models 2–3). We repeated the multivariate analysis with adjustment for the hsCRP level and found that this abolished the associations of visceral and peritoneal fat with LVDD. The effect of the hsCRP level itself remained significant in both models (OR = 2.56, 95% CI = 1.08–4.72, p = 0.023 and OR = 2.70, 95% CI = 1.21–6.69, p = 0.019, respectively; Table 3, model 4 and model 5), indicating that hsCRP might mediate the effect of visceral and/or peritoneal fat on LVDD. To elucidate the influence of EpF, we adjusted for all of the confounding factors and parameters of inflammation and adiposity simultaneously in model 6. EpF remained an independent, significant predictor of the presence of LVDD in this model, which may imply that EpF influences LV diastolic function through pathways other than subclinical inflammation.
Results
Demographics
Of the 149 subjects in our sample population, 65 were diagnosed with LVDD, and the remainder served as controls. The baseline characteristics of the participants in both groups are summarized in Table 1. Consistent with previous reports on hypertension, the subjects with LVDD were predominantly female, older, and suffered more frequently from hypertension or hyperlipidemia (higher LDL level). Patients with LVDD had significantly higher hsCRP levels (Table 1). The cause of renal failure and the residual renal function did not differ between the groups. Relative to the control group, subjects with LVDD had larger end-diastolic and systolic LV volumes (p < 0.05), greater LA diameters, and larger indexed LV mass values (p < 0.05). Comparison of the functional parameters showed a prolonged deceleration time (DT) (p < 0.05), increased mitral inflow late filling wave (p < 0.001), decreased mitral inflow E/A ratio (p < 0.005), and decreased peak annular early diastolic velocity of the lateral mitral annulus in tissue Doppler imaging (p < 0.001) among the patients with LVDD.
Comparison of the anthropometric characteristics showed higher levels of markers reflecting fat distribution, such as the amounts of total, subcutaneous, visceral, and peritoneal fat (p < 0.001, p < 0.001, p < 0.01, and p < 0.005, respectively), in the LVDD group.
Correlation Between EpF Thickness and LVDD
The bivariate Pearson correlation coefficients for LV diastolic function parameters and EpF are shown in Figure 2. EpF was significantly associated with tissue Doppler e', E/e', and DT (r = −0.39, p < 0.001; r = 0.27, p = 0.001; r = 0.29, p < 0.001, respectively) (Figure 2A–C). EpF thickness was greater in patients with LVDD (n = 65; 5.1 ± 2.6 mm) than in controls (n = 84; 2.8 ± 1.6 mm, p < 0.001, Figure 2D).
(Enlarge Image)
Figure 2.
Correlation between epicardial fat (EpF) thickness and left ventricular diastolic dysfunction (LVDD). EpF thickness correlated significantly with (A) e' (r = −0.39, p < 0.001), (B) E/e' (r = 0.27, p < 0.001), and (C) mitral inflow deceleration time (r = 0.29, p = 0.002). (D) EpF thickness in patients with or without left ventricular diastolic dysfunction (LVDD). The bar graph shows the mean + standard deviation. EpF thickness was significantly greater in patients with LVDD (5.1 ± 2.6 cm) than in controls (2.8 ± 1.6 cm, p < 0.001).
Factors Associated With LVDD
We performed univariate analysis to determine the risk factors associated with the development of LVDD. Hypertension, the natural logarithm of the LDL level [Ln(LDL)], Ln(Visceral fat), Ln(Peritoneal fat), the hsCRP level, Ln(LV mass index) and EpF thickness were significantly associated with the presence of LVDD (Table 2).
Accumulation of EpF Rather Than Visceral Fat is an Independent Risk Factor for LVDD
We previously found subclinical inflammation (e.g., the plasma hsCRP level) to be an independent risk factor for LVDD in patients with metabolic syndromes. We therefore performed multivariate analysis of the risk factors other than adiposity that were associated with LVDD in our univariate analyses and found that the hsCRP level remained associated with LVDD after adjustment for age, gender, diabetes mellitus, hypertension, the hsCRP and LDL levels, and the LV mass index (OR: 1.75, 95% CI: 1.08–3.12, p = 0.038; Table 3, model 1).
We further evaluated fat distribution and its association with LVDD by multivariate regression analysis. Of the CT measures of fat distribution, Ln(visceral fat) and Ln(peritoneal fat) remained associated with LVDD after adjustment for the abovementioned confounding factors (adjusted OR = 2.29, 95% CI = 1.06–5.57, p = 0.013 and adjusted OR = 2.45, 95% CI = 1.1–5.9, p = 0.011 respectively; Table 3, models 2–3). We repeated the multivariate analysis with adjustment for the hsCRP level and found that this abolished the associations of visceral and peritoneal fat with LVDD. The effect of the hsCRP level itself remained significant in both models (OR = 2.56, 95% CI = 1.08–4.72, p = 0.023 and OR = 2.70, 95% CI = 1.21–6.69, p = 0.019, respectively; Table 3, model 4 and model 5), indicating that hsCRP might mediate the effect of visceral and/or peritoneal fat on LVDD. To elucidate the influence of EpF, we adjusted for all of the confounding factors and parameters of inflammation and adiposity simultaneously in model 6. EpF remained an independent, significant predictor of the presence of LVDD in this model, which may imply that EpF influences LV diastolic function through pathways other than subclinical inflammation.
Source...