Epicardial fat tissue can predict subclinical left ventricular dysfunction in patients with erectile dysfunction

Abstract Background Erectile dysfunction (ED) is an early form of atherosclerosis and subclinical myocardial dysfunction. Epicardial fat tissue (EFT) is associated with impaired left ventricular (LV) function, even in the absence of cardiovascular disease. The aim of this study was to investigate the association between EFT and LV systolic function in patients with erectile ED by speckle tracking echocardiography (2D-STE) method. Methods A total of 129 consecutive patients with ED were compared with 145 age- and sex-matched control subjects. ED was evaluated using the International Index of Erectile Function questionnaire. Thickness of EFT was measured by TTE. Global LV longitudinal strain (LV-GLS) and global LV circumferential strain (LV-GCS) were measured by 2D-STE method. Results The EFT thickness was significantly higher in the patients with ED (p <.01). LV-GLS and LV-GCS were revealed to be more deterioration in the ED group compared to controls (−18.2 ± 2.7 vs. (−21.1 ± 3.9, p<.001; −19.5 ± 4.1 vs. −21.9 ± 3.9, p<.001, respectively). It has been shown that EFT thickness is an independent predictor of LV dysfunction. Conclusions These results indicate that EFT thickness is associated with subclinical LV systolic dysfunction in patients with ED.


Introduction
Erectile dysfunction (ED) is not a simple sexual activity disorder by itself, but by definition is an early form of atherosclerosis [1]. Similar to atherosclerotic diseases, ED patients have risk factors such as diabetes mellitus (DM), hypertension (HT), dyslipidemia, smoking, and metabolic syndrome [1,2]. Endothelial cell damage and endothelial dysfunction caused by oxidative stress constitute the basic pathogenesis of the disease [1,2]. The emergence of symptoms of ED approximately 2-5 years before the onset of various cardiovascular diseases, and the determination of a relationship between the severity of ED and the severity of cardiovascular events, shows that ED is a leading condition for the development of atherosclerosis [3,4]. However, ED has recently been recognized as an independent indicator of cardiovascular disease risk. Therefore, the presence of ED may provide an opportunity for early detection of cardiovascular disease risk in men without overt cardiovascular disease [3,4].
Epicardial fat tissue (EFT) is a metabolically active tissue surrounded by the visceral pericardium that also produces inflammatory, atherogenic cytokines, and is considered to be an indicator of increased cardiovascular risk in many clinical situations [5][6][7]. Studies have reported that the increase in EFT thickness is associated with subclinical atherosclerosis, coronary artery disease (CAD), left ventricular (LV) dysfunction, and metabolic syndrome [8][9][10][11].
Previous studies have shown that increased EFT is associated with impaired LV function, even in the absence of cardiovascular disease [12][13][14]. At the same time, increased EFT thickness has begun to be accepted as a cardioembolic risk factor [5,6,8]. Studies conducted with ED patients have found a relationship between ED and EFT thickness, and it has been shown that the thickness of EFT in these patients is proportional to the current cardiovascular risk status [15,16].
Two-dimensional speckle tracking echocardiography (2D-STE) is a new imaging method used to evaluate cardiac mechanics [17]. Myocardial function analysis with 2D-STE is superior to many other conventional imaging techniques [17]. It is extremely valuable in defining subclinical myocardial dysfunction in various clinical situations, and early detection of these subclinical changes in myocardial tissue will provide an opportunity for early diagnosis and treatment of cardiovascular complications [18]. Further evaluation of cardiovascular risk is recommended for men diagnosed with ED using non-invasive methods to detect subclinical cardiovascular disease [3,4]. Therefore, the aim of this study is to investigate the value of EFT in predicting LV subclinical dysfunction in patients with ED with normal LV ejection fraction in conventional 2D echocardiography and without overt cardiovascular disease, using 2D-STE method.

Study population
A total of 126 consecutive patients admitted to the urology outpatient clinic with a diagnosis of ED and planned to determine cardiovascular risk were included in the study. A total of 132 age and gendermatched volunteers were selected as the control group. Those with LV segmental motion defects, history of CAD (history of percutaneous coronary intervention and coronary bypass), DM, LV ejection fraction <55%, previous cerebrovascular disease, peripheral artery disease, conduction abnormalities, atrial fibrillation, valvular heart disease more than mild, any cardiac surgery history of pacemaker, polyneuropathy due to surgical trauma (radical retropubic prostatectomy, cystectomy, etc.), neurological diseases, poor echocardiographic image and positive cardiovascular stress test, phosphodiesterase inhibitors and the patients taking beta blockers were not included in the study. Traditional risk factors such as HT, age, smoking, and family history of CAD of all participants was recorded. Cardiovascular stress test was applied to all participants using Bruce Protocol to investigate ischemia. A 12-lead ECG with a filter range of 0.5-150 Hz (25 mm/s, 10 mm/mV) was performed in all patients. Blood samples were taken to evaluate routine biochemical and hematological parameters after 12 h of fasting. A systolic blood pressure value of 140 mm Hg and/or a diastolic blood pressure value of !90 mmHg was defined as HT. Informed consent was obtained from all participants and the study was approved by the local ethics committee.
Transthoracic echocardiogram (TTE) examination of all participants was performed in left lateral lying position using 2.5--3.5 MHz ultrasound probe (Philips Affiniti 50, Amsterdam, Netherlands). Standard echocardiographic measurements such as left atrium length, LV systolic and diastolic diameters, LV wall thicknesses and LV ejection fraction (LVEF) were performed in accordance with the European Society of Cardiology Echocardiography guidelines [19].

Imaging and measurement of epicardial fat tissue
Echocardiographic assessment of EFT thickness was performed as described by Iacobellis and Willens [5]. The measurement of EFT thickness was performed by TTE from a parasternal long-axis view on the right ventricle's free wall at end-diastole, and the greatest perpendicular distance to the aortic annulus was achieved and averaged over three cardiac cycles [5]. In the parasternal long-axis window, hypoechoic space on the right ventricular free wall was defined as EFT ( Figure 1) [5]. EFT was detected as an area of relatively low echogenicity located between the right ventricle and the inner leaf of the pericardium. In order to detect intra-observer variability, 25 individuals were randomly selected, and the EFT thickness of these 25 individuals was measured once every week. Accordingly, the repeatability of EFT thickness measurement was calculated with the intraclass correlation coefficient analysis. It was observed that the repeatability of EFT thickness measurement was quite good (intra-class correlation coefficient ¼ 0.915; p<.001) [3]. The diagnosis of ED was made according to the answers given to five questions in the International Index of Erectile Function (IIEF-5) and was categorized according to the total score as stated previously [20]. Those with a total score of 22-25 were determined as non-ED, those with a total score of 17-21 as mild ED, those with a total score of 12-16 as mild-moderate ED, those with a total score of 8-11 as moderate-ED and those between 1 and 7 as severe ED [20].

Speckle tracking echocardiography
STE and GLS were assessed from standard two-dimensional, gray-scale images derived from apical two-three and four-chamber views. GCS was evaluated using the parasternal short-axis images in LV basal, mid, and apical levels. The images were obtained during an end-expiratory breath-hold at a frame rate of 60-80 frames/s. For further offline analysis, all images were transferred to a workstation and automated software QLAB quantification software version 7.1 (Philips, the Netherlands) was used. Automated tracking of myocardial speckles was then reviewed and manually adjusted as minimally as possible. The endocardial border was manually traced in the end-systolic frame for each view. The entire circumference of the LV was divided into six equal segments by the software. Myocardial strain curves were then generated via the software by frame-by-frame tracking of the natural acoustic markers throughout the cardiac cycle. Longitudinal strain was calculated from the apical two-chamber, three-chamber, and four-chamber views, whereas circumferential strain was derived from the three short-axis views. Fifteen patients with inadequate image quality for GLS and GCS analysis were excluded from the study. Also, global systolic strain rate (GSRs), the global diastolic strain rate during the early (GSRe), and late (GSRa) phase of diastole were analyzed.

Statistical analysis
All measurements were evaluated by Kolmogorov-Smirnov test in terms of compliance with normal distribution. Continuous variables were given as mean ± standard deviation, and categorical variables as frequency (percentage). In comparison of groups, student-t or Mann-Whitney U test was used for continuous variables, and chi-square test for categorical variables. Spearman correlation analysis was performed to determine the correlation between values. Multiple linear regression analysis was used to determine independent predictors of LV systolic functions. Statistical significance was defined as p<.05. SPSS version 22.0 (SPSS 22.0 for Windows, Inc., Chicago, IL) was used for all statistical calculations.

Results
The demographic, clinical characteristics and laboratory parameters of the patients and the control group are given in Table 1. There was no difference in the frequency of HT, smoking, and family CAD in both groups. In addition, there was no difference between the two groups in values such as glomerular filtration rate (GFR), body mass index (BMI), and age. For laboratory values other than LDL, there was no difference between the two groups, while the LDL value was significantly higher in the ED group.
Among conventional echocardiography parameters, no difference was observed between LVEF, left atrium diameter, LV wall thickness, left ventricular mass index (LVMI), ventricular systolic and diastolic diameters ( Table 2). While the E/A ratio was similar in both groups, E/e' was higher in the ED group (6.8 ± 2.2 vs. 8.4 ± 2.6, p¼.012). EFT thickness was observed to be significantly higher in the group with ED (p < 0.001) ( Table 2). When Spearman correlation analysis was performed, a significant correlation was observed between EFT thickness and IIEF-5 score (r¼ À0.485, p<.001) (Figure 2).
Comparison of LV strain and strain rate between groups is shown in Table 3. GLS and GCS are negative values due to shortening of a myocardial segment in each direction. GLS and GCS were revealed to be more deterioration in the ED group compared to controls (À18.2 ± 2.7 vs. À21.1 ± 3.9, p<.01; À19.5 ± 4.1 vs. À21.9 ± 3.9, p <.01, respectively) ( Table 3).
Longitudinal and circumferential systolic strain rate was lower in the ED group compared to controls (-1.33 ± 0.41 vs. À1.81 ± 0.22, p<.001; À1.41 ± 0.35 vs. À1.72 ± 0.24, p <.001, respectively). There were no significant differences with longitudinal early and late diastolic strain rates between the ED and control groups (p>.05) ( Table 3). There was also no significant difference circumferential early and late diastolic strain rates between the ED and control groups (p>.05) ( Table 3). A multivariate linear regression analysis was performed to determine the independent predictors of GLS and GCS. Age, HT, IIEF-5 score, and EFT thickness were independently associated with GLS (Table 4). Also age, E/e', IIEF-5 score, and EFT thickness were independently associated with GCS (Table 5).

Discussion
In this study, for the first time, we investigated the association between EFT and subclinical myocardial systolic dysfunction in patients with ED by 2 D-STE method. Our results showed that ED is associated with subclinical deterioration on the left ventricle systolic function. Even if the absence of overt cardiovascular disease and risk factors, GLS and GCS were detected to be lower in patients with ED and also there was a strong correlation between ED severity and GLS and GCS. It also showed that EFT thickness was higher in ED patients than the control group and that there was a relationship between the severity of ED and EFT thickness. In addition, we found that EFT was independently associated with both GLS and GCS in ED patients. These findings clearly demonstrated that EFT thickness in ED patients is associated with subclinical LV systolic dysfunction.  Figure 2. The Spearman correlation analysis showed that a significant correlation was observed between EFT thickness and IIEF-5 score.  ED is a common health problem that is increasing in frequency today and affects quality of life [20]. It has been shown that ED is no longer a simple sexual disease, but an early predictor of cardiovascular diseases and mortality [21]. Previous studies focused particularly on the fact that ED is a precursor of CAD [21]. It has been shown that cardiovascular events generally occur 2-5 years after the diagnosis of ED [22]. In a study, it was shown that 57% of patients with a history of coronary bypass had ED in their history [23]. It has been shown that the presence of ED may also be the first sign of systemic vascular disease [24,25]. Previously, ED was considered to be a result of systemic disorders such as HT, DM and other vascular diseases, but now the common view is that ED is an early sign of early atherosclerosis and thus systemic vascular diseases [25]. The above studies support this view. Subclinical endothelial dysfunction and inflammation are thought to be the main pathophysiological factors in the occurrence of cardiovascular diseases in ED patients [26,27]. Low-grade subclinical inflammation in ED may lead to an atherosclerotic process by causing deterioration in endothelial function [26,27]. It has been clearly demonstrated that damage to the penile artery occurs before the development of clinically significant atherosclerotic disease [26]. Recently, Gandaglia et al. stated that ED and cardiovascular diseases should be seen as two different symptoms of the same systemic condition, and that ED is an early marker of symptomatic cardiovascular diseases [27]. In support of the hypothesis that ED may serve as a marker of silent cardiovascular disease, Jackson et al. Detected CAD by computed tomographic coronary angiography in approximately half of men presenting with ED but without cardiac symptoms [28]. The Third Princeton Consensus Conference suggested further cardiac evaluations, pointing out that ED patients without known cardiovascular disease should be considered at increased risk of cardiovascular disease [29].
In this study, for the first time, we aimed to determine the role of EFT in predicting subclinical myocardial systolic dysfunction in ED patients with normal conventional echocardiographs, with 2 D-STE method.
Two-dimensional speckle tracking echocardiography (2 D-STE) is an effective method used to assess myocardial systolic and diastolic function in detail [30] Moreover, it is a useful method to evaluate Twodimensional speckle tracking echocardiography (2 D-STE) and is an effective method used to assess myocardial systolic and diastolic function in detail [30]. Moreover, it is a useful method to evaluate subclinical LV systolic dysfunction in patients with normal ejection fraction on conventional echocardiography [30].
In our current study, we found that although EF was normal in conventional ECHO, there was a decrease in GLS and GCS values measured with 2 D-STE and subclinical LV systolic dysfunction in patients with ED. Similarly, previous studies have shown that LV systolic dysfunction is more common in ED patients compared to control groups [31]. In a recent study, Zehir et al. found that while EF was normal in ED patients, GLS and GCS values decreased [32]. In addition, Karag€ oz et al. found decreased GLC and GLS values in ED patients even after excluding cardiovascular risk factors in their study [31]. Our study is also considerable in terms of showing the relationship between LV systolic dysfunction and ED degree. In our study, as the ED level increased, we observed a decrease in the GLS and GCS values. As the ED level increases, the increase in endothelial dysfunction at the same degree may explain these results. Uslu et al. showed that endothelial dysfunction is associated with the progression of myocardial systolic dysfunction, suggesting that endothelial dysfunction developing in ED patients may play a role in myocardial deterioration [33]. However, microvascular dysfunction due to cumulative coronary risk factors leading to endothelial dysfunction may occur without significant coronary stenosis and may cause contractile abnormalities [33]. Another recent study showed that the LV function and endothelial dysfunction deteriorate in patients with ED without evident CAD [34]. These findings can help explain the deterioration of GLS and GCS values, even in the absence of overt DM and CAD. The fact that it is possible to detect these subclinical changes in the ventricle with global deformation analysis shows the importance of the test [33].
EFT is a metabolically active tissue surrounded by the visceral pericardium that also produces pro-inflammatory, pro-atherogenic cytokines, and is considered to be an indicator of increased cardiovascular risk in EFT: epicardial fat tissue; E/e': ratio between early diastolic mitral inflow velocity and early diastolic annular velocity; IIEF-5: International Index of Erectile Function-5; HT: hypertension; LDL: low-density lipoprotein; LV-GCS: global circumferential strain many clinical situations [5][6][7]. EFT can affect ventricular function through impaired microvascular relaxation by secreted adipokines or direct toxic effect on myocardium [33][34][35]. Previous studies have shown that increased EFT is associated with impaired LV function even without CAD [10,31,36]. In another study, Konishi et al. found that the increased volume of EFT detected by computed tomography (CT) was associated with LV dysfunction, regardless of other general risk factors such as age, gender, diabetes, HT, or abdominal obesity [10]. Previous studies have shown that there is an independent relationship between EFT and myocardial systolic function [35,37]. It has also been shown that the increase in EFT thickness correlates with LV dysfunction [35]. A recent in-vitro study showed that EFT from guinea pigs secreted factors that inhibit cardiomyocyte contractile function and induce insulin resistance [36]. In studies conducted with ED patients, a significant relationship was found between visceral adipose tissue in various parts of the body and ED [15,16]. In a study, a significant relationship was found between the presence of ED and EFT, and EFT was found to be associated with other cardioembolic risk factors in this patient group [16]. The similarity of risk factors that cause ED development and increase in EFT thickness increases the strength of our study [15,16]. In our study, we detected significantly higher EFT thickness in ED patients compared to the control group. The IIEF-5 test has been shown to be a very useful method used in clinical practice to identify men with ED and to rate ED levels [20]. In this study, we also think that the significant correlation between EFT thickness and IIEF-5 strengthens the close relationship between EFT and ED. In other words, the fact that patients with severe ED have higher EFT thickness than mild ED patients indicates that these patients require more stringent monitoring for CAD. It may explain the relationship between ED and EFT increase, because they share common pathophysiological pathways and risk factors, including endothelial dysfunction [21]. Both EFT and ED have been shown to be associated with LV systolic dysfunction [11,34]. These previous studies clearly show the relationship between ED and LV systolic dysfunction, as well as the relationship between EFT and LV systolic dysfunction [11,34]. Confirming these relationships in the same patient population makes our current study stronger. In this study, we determined that EFT thickness was independently associated with decreased GLS and GCS in ED patients. A recent study showed that the presence of coronary microvascular dysfunction is an independent marker of ED in men without CAD [28]. Similarly, a relationship was found between increased EFT and microvascular dysfunction developing in coronary arteries [38]. Therefore, it has been revealed that the developing microvascular disorder can play an important role both in the presence of increased EFT and in the development of ED [28]. In conclusion, it is not surprising that the presence of increased EFT in ED patients predicts deterioration in ventricular functions. In our study, it is not difficult to explain the high EFT thickness and impaired LV systolic functions in ED patients with the above mechanisms.
In addition, diabetic patients and in patients with CAD or peripheral artery disease were not included in our study to show the absolute effects of ED on LV function. Therefore, we think that the presence of ED may cause subclinical impairment in LV systolic function. These findings suggest that, even in the presence of normal LV systolic function in conventional echocardiographic evaluation, ED patients should be made more strait cardiovascular risk stratification using more advanced imaging methods, and the hypothesis that a more detailed assessment of cardiovascular disease is required in the presence of ED in men over 50 years old seems to be confirmed by the results of our study [39]. Therefore, it seems reasonable to note the higher EFT thickness in the echocardiography of patients with ED. Our results showed that high EFT thickness may have affected LV systolic function in patients with ED. In light of these findings, we suggest that EFT can be used to detect individuals with increased cardiovascular risk in ED patients.

Limitations
Our findings should be considered in the light of potential study limitations. The first limitation can be considered as the low number of patients, especially the small number of patients in subgroups. In our study, we excluded major cardiovascular risk factors such as overt CAD, DM, and peripheral vascular disease, but there may be silent CAD in study subjects, since there are studies of ED patients without cardiac symptoms but with CAD detected by CT. Although there is a significant relationship between increased EFT thickness and LV dysfunction, this cause-effect relationship cannot be fully explained. Although epicardial fat accumulation can be measured most accurately using MRI or CT imaging, it was observed that the results in studies performed with echocardiography coincided with CT or MRI studies and it was stated that echocardiographic EFT measurement could be used [3]. Finally, we did not have follow up data of patients.

Conclusions
Our study results revealed that there is a decrease in GLS and GCS values in patients with ED, that is, impairment in LV systolic functions. In patients with ED, a more comprehensive cardiovascular risk classification is required even in the absence of main major cardiovascular risk factors or overt cardiovascular disease. The STE parameters can identify subtle changes before clinical signs of a cardiovascular disorder occur. STE parameters can identify ED patients at high risk of developing cardiovascular disease before clinical signs of a cardiovascular disease occur. The EFT thickness was higher in the patients with ED. The ED patients with higher EFT thickness had more subclinical LV deterioration. We suggest that EFT thickness may be used to predict future cardiovascular events in patients with ED.