Epicardial adipose tissue is a predictor of ascending aortic dilatation in hypertensive patients, but not paracardial adipose tissue

Background Ascending aortic aneurysms are one of the major causes of mortality. In recent years, there is a growing interest of epicardial adipose tissue (EAT) and related diseases. The aim of this study was to investigate the relationship of EAT, and PAT with ascending aortic dilatation (AAD). Methods We included 934 patients with hypertension in this study. The patients were evaluated by a complete transthoracic echocardiographic examination, including measurements of EAT, PAT, and aortic dimensions. Aortic size index (ASI) was used for diagnosing AAD. The patients were divided into two groups: dilated ascending aorta diameter (ASI ≥ 21 mm / m2, n = 102) or normal aortic diameter (ASI < 21 mm / m2, n = 832) according to the ASI. Characteristics of these patients were compared before and after propensity score matching analysis. Results Patients with AAD were older (72.3 ± 11.6 vs. 61.7 ± 12.7 years, p <  0.001), had more female gender (66% vs. 54%,p = 0.021) than patients with normal ascending aorta (AA). After propensity score matching analysis (77 vs. 77), EAT [OR:1.461, %95CI (1.082–1.974), p = 0.013] was independently associated with AAD in multivariate logistic regression analysis. In ROC curve analysis, EAT > 0.45 cm had 51.9% sensitivity and 62.3% specificity [AUC = 0.617, P = 0.012, 95% CI (0.529–0.707)]. Conclusion Based on our findings, increased EAT may be suggested as an independent risk factor for AAD due to local or systemic effects in hypertensive patients.


Background
Ascending aortic aneurysms and dissections are one of the major causes of mortality [1]. Despite important recent development in understanding of its pathogenesis, the role of haemodynamics and other reasons are partially known yet. The pathogenesis of ascending aortic dilatation (AAD) includes several factors with systemic and local effects. The aetiology is multifactorial, including environmental and genetic factors that play important roles in progression of aortic disease [2]. Given that the pathogenesis of AAD maintains unclear, the main reason seems to be related to the deficiencies or defects in structural proteins, such as collagen and elastin in the aortic wall [3]. Changes of the extracellular matrix are suggested as a key factor in the pathogenesis of AAD [2][3][4]. Cystic medial degeneration responsible for weakened aortic wall is related to the degeneration of smooth muscle cell and elastic fibre apoptosis [5]. Also, the inflammatory process associated with atherosclerosis causes aortic dilatation [6].
Hypertension is one of the considerable risk factors for AAD. AAD is often observed in hypertensive patients compared with normotensive individuals and is correlated with cardiac and noncardiac organ injury in hypertensive patients [2][3][4]. Hypertension is one of the most common diseases and a worldwide health problem. It is related to stroke, renal disease, cardiac disease, aortic disease, mortality and morbidity. It generally progresses silently until it causes end organ damage. However, it is a treatable and preventable disease.
In recent years, there is a growing interest towards the effects of epicardial adipose tissue (EAT) on hypertension and cardiovascular risk factors. EAT is a visceral adipose tissue found between the pericardium and the myocardium [7]. It is generally placed in the atrioventricular and interventricular groove throughout the coronary arteries. There is no any fascia between the myocardium and EAT; therefore, they have same microcirculation. EAT secretes antiinflammatory and antiatherogenic mediators and supplies energy to the myocardium [8][9][10]. However, under pathological conditions, EAT appears to play a critical role in atherosclerosis, hypertension and progression of metabolic diseases as it acts as a proinflammatory and prothrombotic organ [11]. Tumour necrosis factor alpha, interleukin-6 and monocyte chemoattractant protein-1, secreted from the EAT [12], immediately affect atherosclerosis [13,14]. As a result, increased EAT is related to atherosclerosis [15]. EAT may play a provocative role for AAD with paracrine and systemic endocrine effects on vessels [16].
Paracardial adipose tissue (PAT) is an ectopic fat depot surrounding the heart [17,18]. It is placed anterior to the EAT and outside the parietal layer of the pericardium [19]. The pericardium restricts the communication between the PAT and EAT. Studies about the different roles between EAT and PAT are limited. A few studies showed that PAT was more closely related with abdominal visceral adipose tissue and metabolic risk than EAT. It is also recommended that PAT might be a marker for increased metabolic risk and visceral adipose tissue on thoracic imaging [17][18][19][20][21].
The focus of the studies has been commonly on EAT, neglecting the potentially additive effects by the PAT on cardiovascular diseases. The aim of this study was to investigate the associations of EAT, and PAT with AAD in hypertensive patients.

Methods
The present study included 934 hypertensive patients who underwent complete transthoracic echocardiographic examination involving measurements of aortic dimensions, EAT and PAT. When diagnosing AAD, we used the body surface area (BSA)-adjusted classifications. For this purpose, we used Roman's classification (aortic size index, ASI) [22]. ASI was calculated by the following formula: ASI = AA diameter (mm)/ BSA (m 2 ) [23]. AAD was defined as a ASI ≥ 21 mm/m 2 according to the Roman's classification [23]. A total of 102 hypertensive patients with AAD were compared with 832 hypertensive patients with normal aortic diameter. Exclusion criteria were bicuspid and rheumatic aortic valve disease, Marfan syndrome, Ehlers-Danlos syndrome, Familial Thoracic Aortic Aneurysm Syndrome, Turner Syndrome and other connective tissue disorders, infectious conditions, restrictive and hypertrophic cardiomyopathy, renal failure requiring dialysis, malignancy and pregnancy. Hypertension was defined as the documentation of a blood pressure of more than 140/90 mmHg or the use of antihypertensive drugs.
The study was approved by the Institutional Ethics Committee and conducted in accordance with the principles set out in the Declaration of Helsinki.
Transthoracic echocardiography was carried out using a Vivid S5 GE Healthcare system. Each patient underwent two-dimensional transthoracic echocardiography according to the recommendations of the European Association of Echocardiography [24]. Parasternal long axis view was used to view the proximal AA. The ascending aortic diameters were calculated between the inner edges of the aortic lumen perpendicular to the long axis 2 cm above the sinotubular junction at the end of the diastole in views showing the largest aortic diameter [16]. The parasternal long axis view was performed to measure maximal EAT thickness. EAT was defined as echo-free space between the outer wall of the myocardium and the visceral layer of the pericardium [14]. PAT was defined as the hypoechoic space in front of EAT and on the external pericardium [14][15][16][17][18][19][20][21][22][23][24][25]. The values were measured in three cardiac cycles and were averaged.

Results
A total of 934 hypertensive patients were divided into the following two groups: patients with ascending aortic dilatation (AAD) (102 patients) and patients with normal ascending aorta (AA) (832 patients). Baseline characteristics and echocardiographic and biochemical parameters are summarised in Table 1. Patients with AAD were older than the patients with normal AA (72.3 ± 11.6 vs. 61.7 ± 12.7 years, p < 0.001), had more female gender (66% vs. 54%,p = 0.021) compared with patients with normal ascending aorta (AA). BMI and BSA were lower in patients with AAD than patients with normal AA (Table 1). EAT thickness (0.48 ± 0.15, p < 0.001) was significantly higher in the patients with AAD compared with the patients with normal AA. There was no difference between groups regarding PAT thickness (0.49 ± 0.18 vs 0.49 ± 0.17, p = 0.738). β-Blocker usage is higher in patients with AAD than patients with normal AA.

Discussion
In this study, we found that EAT are independent predictor for AAD in hypertensive patients. The normal diameter of the AA depends on the patient's age, sex and body size [26]. Similarly to our study, Bon et al. showed that higher age was associated with larger descending aortic diameters. The causes of aortic aneurysms differ according to the location of the aorta. In younger patients, thoracic aortic aneurysms with genetic reasons generally include the AA and aortic root. These reasons include connective tissue disorders, such as Ehlers-Danlos Syndrome, Marfan Syndrome, Loeys-Dietz Syndrome, Familial Thoracic Aortic Aneurysm Syndrome, Turner Syndrome and bicuspid aortic valve. Also, cystic medial degeneration can be seen in thoracic aortic aneurysms that are not due to a connective tissue disorder but have a genetic connection. Other reasons of thoracic aortic aneurysms are aortitis (syphilis is the most common cause) or inflammation of the aorta. A total of 39% of the variance in descending aortic diameter and 21% of the variance in ascending aortic diameter were explained by sex, age, traditional cardiovascular risk factors and anthropometrics [27]. Consequently, only a small part of the causes of ascending aortic aneurysm could be explained by conventional risk factors. The influence of these cardiovascular risk factors seems of limited importance in elucidating the overall reasons for aortic dilatation.
EAT may help fill this gap. There are two possible mechanisms for the association between EAT and AAD. First, EAT has a paracrine and systemic endocrine role. It can secrete several active molecules, such as tumour necrosis factor, resistin, adiponectin and IL-6 [28,29]. Second, EAT is one of the parts of the visceral adipose tissue, and, therefore, it is associated with CAD, cardiovascular   [30][31][32]. These specifications of EAT may be a possible causative role of AAD.
In the literature, PAT and EAT have been used interchangeably, because they were thought to be identical adipose depots. However, paracardial and epicardial adipose tissues are clearly different embriologically, physiologically, anatomically and clinically. Despite this, the literature frequently contradictorily discriminates between the two adipose tissues [33]. EAT is the fat depot placed between the myocardium and visceral pericardium, whereas PAT is the fat depot outside the pericardium and on the external surface of the parietal pericardium as defined by imaging studies and autopsies . So, EAT is the adipose tissue depot immediately adjacent to the myocardium, whereas PAT is the outer adipose tissue of the heart.
EAT originates from the splanchnopleuric mesoderm, whereas PAT originates from the thoracic primitive mesenchyme. EAT is supplied by the coronary arteries, whereas PAT is supplied by the noncoronary system. Also, no any fascia separates the EAT and coronary arteries; therefore, EAT and myocardium share the same circulation. Because of this link, EFT is metabolically an active tissue and the source of cytokines and interacts directly through vasocrine and paracrine mechanisms to the myocardium, coronary and systemic circulation [34]. This is not true for PAT. PAT does not directly interact with the myocardium and may only contact indirectly through acts like ectopic visceral adipose tissue. As a result, pericardial and EATs are clinically different [35].
CAD is correlated with AAD in univariate regression analysis, but this correlation was not determined in multivariate regression analysis after matching. Atherosclerosis can lead to aortic dilatation, especially in aortic arch and descending aorta. Unlike dilatation of the descending aorta, the role of atherosclerosis in the aetiology of AAD is controversial, and AAD is not generally due to atherosclerosis. Most aetiology of the AADs are classified as idiopathic [36,37].
Interestingly, the presence of DM was related to a smaller ascending and descending aortic diameter in the literature [38]. This phenomenon might be caused by high glucose levels associated with DM which inhibits intermediate-phase secretion of the matrix metalloproteinases. Similarly, protective effect of DM has already been shown in the abdominal aorta [39,40]. However, this phenomenon may be more remarkable in abdominal aorta than AA. We did not find any relationship between DM and AAD.
It has been shown that there was a relationship between EAT density and arterial inflammation and cardiovascular risk prediction. This adipose tissue-arterial wall association may play a role in the formation of AAD [41]. The findings of this study may support the hypothesis that increased EAT, but not PAT, may have a critical role in AAD.

Conclusion
Only a little part of the aetiology of AAD was explained by age, sex, traditional cardiovascular risk factors and anthropometrics. EAT may have a key role in AAD. Based on our findings, increased EAT, and BMI were independent predictors for AAD in hypertensive patients. Increased EAT may be effective in progressing aortic dilatation due to local or systemic effects. However, this relationship was not observed in PAT. We believe that further studies are needed to explain the role of different types of adipose tissue in the pathogenesis of AAD. Adipose tissue-focused diagnosis and treatments can provide new insight to the patients with AAD.

Study limitations
This was primarily a cross-sectional study retrospective in nature. MRI and computed tomography are frequently used in the evaluation of adipose tissue. Although MRI is the modality of choice for water/fat separation, CT is considered to gold standard for volumetric adipose tissue analyses. However, we used echocardiography to measure EAT. Also, there are some limitations to the measurement of EAT and PAT by echocardiography. Echocardiographic imaging only allows for a rough estimation of the adipose tissue [42,43]. The restricted acoustic window limits the imaging of total adipose tissue volume and determining regional differences in adipose tissue distribution. Nevertheless, there is a close correlation between MRI and echocardiography imaging of EAT [44,45]. In our study, predictive value of EAT for AAD was modest. The cutt-off values of epicardial fat for AAD prediction are not well defined in literature. Whether or not echocardiographic epicardial fat thickness may really have the diagnostic properties to serve as an indicator of AAD should be analyzed in large, randomized studies.