Introduction

Coronary artery diseases (CAD) have become a serious epidemic reason of morbidity and mortality. Numerous studies are investigating many parameters as a marker for coronary artery diseases. One of the main concerns of the cardiology studies is how to prevent coronary artery disease development and how to estimate the risk of the disease before it becomes clinically symptomatic [1].

Variation in the prevalence of non-alcoholic fatty liver disease (NAFLD) is found based on factors such as race, ethnicity, age, gender, geographical distribution, and the diagnostic modality used [2]. NAFLD has been firmly linked to insulin resistance, type 2 diabetes mellitus, dyslipidemia, and metabolic syndrome, especially in patients with sedentary lifestyles, changing dietary patterns, and increased obesity [3]. These factors also comprise the risk profile for CAD. Both CAD and NAFLD share the same underlying pathophysiological mechanism, risk factors, lifestyle modification, and treatment plans and were therefore hypothesized to be closely related [4]. NAFLD is recently suggested to be an independent risk for cardiovascular diseases and is also associated with higher cardiovascular mortality [4].

A result of the studies emphasized that there was some parallelism between visceral adiposity and ischemic heart disease [5]. As both progression of NAFLD and increasing epicardial adipose tissue (EAT) mass have the same risk factors, therefore these two conditions might be associated with each other [6].

The aim of our work was to determine the association between non-alcoholic fatty liver and increased epicardial adipose tissue mass and its relation to severity of CAD.

Methods

A cross-sectional study

The design of the study was approved by the ethical committee of the faculty of medicine, Ain Shams University, and was following the 1975 Helsinki declaration. All patients included in our study have given written informed consent after explaining the procedures and before they participated in the study.

This was a cross-sectional study including 60 patients who had signs and symptoms of coronary artery disease who attended elective coronary angiography to rule out coronary artery disease. Patients with any form of chronic liver disease as alcoholic or viral liver disease, Wilson's disease, and hereditary hemochromatosis and patients with poor echogenicity, pericardial effusion > 0.5 cm, calcified pericardium or pericardial thickness > 5 mm all had been excluded.

Full history, full clinical examination, and measuring of body mass index (BMI) were done to all patients. Laboratory investigations including alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase, gamma GT, fasting sugar, HbA1C, and lipid profile were obtained from all patients.

Calculation of NAFLD fibrosis scoring was also done in form

  1. (1)

    NAFLD fibrosis scoring system (NFS)

    $$1.675+0.037\times\mathrm{age}\;(\mathrm{years})+0.094\times\mathrm{BMI}\;(\mathrm{kg}/\mathrm m^2)+1.13\times\mathrm{IFG}/\mathrm{diabetes}\;(\mathrm{yes}=1,\;\mathrm{no}=0)+0.99\times\mathrm{AST}/\mathrm{ALT}\;\mathrm{ratio}-0.013\times\mathrm{platelet}\;(\times109/\mathrm l)-0.66\times\mathrm{albumin}\;(\mathrm g/\mathrm{dl})$$

    NFS is >  − 1.455 (proved to have no or mild fibrosis) or < 0.676 (proved to have advanced fibrosis).

  2. (2)

    FIB-4 scoring

    $$(\mathrm{Age}\times \mathrm{AST }(\mathrm{IU}/\mathrm{l})/\mathrm{platelets count }(\times 109/\mathrm{litre}) \times \surd \mathrm{ALT }(\mathrm{IU}/\mathrm{l}))$$

FIB-4 score is < 1.30 (proved to have no or mild fibrosis) or > 2.67 (proved to have advanced fibrosis).

Study procedures

Coronary angiography

Selective coronary angiography was performed by 6F or 7F catheters via standard Judkin’s technique using the right or left femoral or radial approaches. Coronary artery lesions above 50% lesion were considered significant while those below 50% lesion were considered non-significant.

Abdominal ultrasound

Liver ultrasonography performed by experienced radiologists correlated to clinical details (Toshiba real-time device using a 3.5 MHz convex transducer). Diagnosis of NAFLD based on detection of the following findings; liver parenchyma hyperechogenicity, poor visualization of intra-hepatic vessel borders, and attenuation of ultrasound beam Semi-quantitative assessment of the severity of hepatic steatosis was also done as follows: 0 (none), grade 1 (mild), grade 2 (moderate), and grade 3 (severe), depending on the discrepancy degree of amplitude of echo between liver and kidney and echoes loss from the intrahepatic vessels walls and diaphragm.

Echocardiography

Trans-thoracic echocardiography was performed (GE Vivid 7 device using 3.5 MHz transducer) to measure EAT thickness using the long and short parasternal axes in the left lateral decubitus position. Pericardial fat thickness was measured on the right ventricle free wall and three measurements were taken and averaged.

Statistical analysis

Data were collected, tabulated, and entered into a PC using (SPSS 25). Descriptive statistics were done using mean and standard deviation for numerical data, frequency, and percentage for non-numerical data. The statistical significance of the difference between two study group means was assessed using the Student T test while the statistical significance of the difference between more than two study group means was assessed using ANOVA Test. The relationship between two qualitative variables was examined using the chi-square test and Fisher’s exact test. P value of ≤ 0.05 was considered significant while a P value of ≤ 0.01 was considered highly significant.

This study was approved by the ethical committee of the cardiology department, Ain Shams University under Federal Wide Assurance No.FWA 000,017585.

Results

Our study comprised of sixty patients. They were divided according to the coronary angiography into 2 groups; group A included those having a significant coronary artery disease (above 50% stenosis) and group B comprised of patients having a non-significant coronary artery disease (below 50% stenosis). Forty-four patients were found to be in group A (73.3%) while 16 patients were found to be in group B (26.7%).

Regarding group A, they were 36 (82%) males and 2 (18%) females with a mean age of 49.750 ± 10.607 while in group B, they were 14 (87.5%) males and 2 (12.5%) females with a mean age of 54.625 ± 8.717. Regarding age, sex, and BMI distribution, no statistically significant differences could be detected between both groups (Table 1).

Table 1 Demographics and anthropometric data of the patients
Full size table

In group A, 28 patients had DM (64%), 29 patients were hypertensive (66%), and 40 patients were smokers (91%) while in group B, 12 patients had DM (75%), 12 patients had HTN (75%), and 12 patients were smokers (75%). Regarding the distribution of risk factors of CAD, no statistically significant differences between both groups did exist (Table 2).

Table 2 CAD risk factors distribution in group A versus group B
Full size table

Regarding the hepatic and metabolic profile, a statistically significant higher liver enzyme [ALT, AST, GGT, ALP] among group A patients than in group B (P value 0.006, 0.001, < 0.001, < 0.001, respectively), while regarding the noninvasive scoring system [NFS and FIB-4] for NAFLD diagnosis, there was no significance between both groups (Table 3).

Table 3 Full hepatic and metabolic profile among group A versus group B
Full size table

Regarding the distribution and grading of NAFLD in our study population, it was as follows: 19 patients were described as having no hepatic steatosis (32%), 21 patients were with grade 1 disease (35%), 18 patients were with grade 2 (30%), and 2 patients had grade 3 (3%). A statistically significantly higher percentage of hepatic steatosis (grades 1 and 2) was found among group A patients in comparison to group B (P value < 0.001) (Table 4).

Table 4 NAFLD grading distribution in group A versus group B
Full size table

No statistically significant relationship between the NAFLD fibrosis scoring system of the patients and their sex or their comorbidities could be detected (Table 5).

Table 5 Relationship between NAFLD fibrosis scoring system and sex of the patients and their comorbidities
Full size table

Average EAT thickness in Group A ranged between 6.7 and 9.7 mm while in group B, it ranged between 4.9 and 6.1 mm yielding statistically higher values among group A patients in comparison to those of group B (P value < 0.001) as shown in Table 6.

Table 6 Average EAT thickness in group A versus group B
Full size table

Regarding the risk factors, average EAT thickness was statistically significantly higher among patients with DM and HTN but not among smokers. No statistically significant relationship was found between the average EAT thickness of the patients and their sex (Table 7).

Table 7 Relationship between average EAT thickness and sex of the patients and their comorbidities
Full size table

A statistically significant positive correlation was found between average EAT thickness on one side and BMI (P value = 0.011) and each of the following patients’ metabolic parameters on the other side: alkaline phosphatase (P value = 0.003), gamma GT (P value < 0.001), fasting sugar (P value = 0.019*), HbA1C (P value = 0.032), and LDL (P value = 0.029) (Table 8).

Table 8 Correlations between average EAT thickness and BMI of the patients and their metabolic parameters
Full size table

Also, statistically significant higher EAT thickness values were found among grade 3 hepatic steatosis than in grades 2, 1, or 0 (P value < 0.001). These data are represented in Table 9.

Table 9 Correlation between average EAT thickness and grading of NAFLD among the study population
Full size table

Finally, our study demonstrated the best cutoff value of EAT thickness in the detection of significant CAD among all studied patients as > 6.1 mm, with 100% sensitivity, 100% specificity, 100% PPV, and 100% NPV with an overall accuracy of 100%. This was shown in Table 10 and Fig. 1.

Table 10 ROC curve of EAT thickness in prediction of significant coronary artery disease (above 50% stenosis)
Full size table
Fig. 1
figure 1

The ROC (receiver operating characteristic) curve for the best cutoff value of epicardial fat volume to predict significant coronary stenosis

Full size image

NAFLD fibrosis scoring system cutoff value that shows discrimination between significant and non-significant CAD, at cutoff >  − 2.18, sensitivity was 79.55, specificity was 50, PPV was 81.4, NPV was 47.1, and accuracy was 56%. These data are represented in Table 11 and Fig. 2.

Table 11 ROC curve of NAFLD fibrosis score in prediction of significant coronary artery disease (above 50% stenosis)
Full size table
Fig. 2
figure 2

The ROC (receiver operating characteristic) curve for the best cutoff value of the NAFLD fibrosis scoring system to predict significant coronary stenosis

Full size image

Discussion

Cardiovascular diseases represent the largest cause of death in the world and the greatest proportion of those deaths are due to CAD. Therefore, more sensitive measures for risk stratification and prevention and treatment of coronary vascular disease are highly required [7].

Excess adiposity had been linked to increased cardiovascular risk. A growing amount of evidence suggests that regional adiposity has been associated with the development of adverse metabolic and cardiovascular risk profiles. The major site of fat accumulation is the subcutaneous tissue and this type of fat represents the “good” fat [7].

Recent increased awareness of the NAFLD is deemed essential as it has been shown to correlate with other diseases, in particular atherosclerotic CAD. It has been suggested that NAFLD induces a systemic inflammatory response, increased oxidative stress, insulin resistance, fatty acid toxicity, and endothelial dysfunction; all of which share the same pathophysiology for the development of atherosclerotic CAD [8]. The study aimed to investigate the association between NAFLD and increased EAT mass and their relationship with coronary artery disease severity.

Our study population consisted of sixty patients who presented with signs and symptoms of coronary artery disease eligible for elective coronary angiography to rule out CAD. In our study, 41 patients were with a history of hypertension (68%), 40 patients with a history of diabetes mellitus (66.7%), and 52 patients were smokers (86.7%). All patients were examined by abdominal ultrasound for the diagnosis of NAFLD. 41 patients were detected as having NAFLD (68%) and they were classified as follows: 21 patients were classified as grade 1 (35%), 18 patients with grade 2 (30%), while 2 patients were described as having grade 3 disease (3%). There were 19 patients with no steatosis (32%).

Relation between NAFLD and cardiovascular risk

The present study showed that patients with significant coronary stenosis had statistically significantly higher liver enzymes [ALT, AST, ALP, GGT] than patients without significant coronary stenosis. While there was an insignificant higher NAFLD fibrosis scoring system and FIB-4 among patients with significant coronary stenosis.

Although reports have addressed the relationship between these liver enzymes and risk factors for CVD, there exists discordance in the relationships between specific liver enzymes and CVD. A meta-analysis performed by Fraser et al. [9] has shown that GGT but not ALT is associated with the incidence of coronary heart disease and stroke. Another stratified analysis has demonstrated that ALT is positively associated with stroke but negatively associated with coronary heart disease [10]. On the contrary, Lee et al. [11] showed that elevated ALT and AST are related to CVD mortality. These variations in previous findings may result from different ages, genders, ethnicities, and sample sizes of the studies and also from variable disease conditions.

Also in this study, patients with significant CAD have a statistically significantly higher percentage of NAFLD by ultrasound(grades 1 and 2) in comparison to those with non-significant CAD (P value = 0.005). The results of our study are concordant with Targher et al. in 2008 who reported that NAFLD per se is associated with accelerated Atherogenesis [12].

Moreover, on applying ROC curve analysis, the best discrimination between significant and non-significant CAD in relation to NAFLD fibrosis scoring was found at cutoff value of > -2.18, sensitivity was 79.55, specificity was 50, PPV was 81.4, NPV was 47.1, and accuracy was 56%.

Accumulation of liver fat and elevated liver enzymes are associated with type 2 diabetes and hypertension, both of which are representative risk factors for CVD. Furthermore, on using 18F-fluorodeoxyglucose positron emission tomography (FDG-PET), it was found that NAFLD is associated with vascular inflammation, which may reflect rupture-prone vulnerable atherosclerotic plaques. Both visceral fat accumulation and risk of metabolic syndrome are significantly correlated with GGT and ALT levels [13].

Relation between EAT and cardiovascular risk

Visceral fat perhaps more than total fat is implicated in atherosclerosis. Epicardial adipose tissue (EAT), a non-traditional visceral fat depot modulates coronary arteries through paracrine or vasocrine secretion of bioactive adipokines [14].

In our study, hypertension, diabetes mellitus, and dyslipidemia (represented by LDL-c) were found to have statistically significant higher EAT thickness. These results are consistent with the findings of a study by Kim et al. in 2015 which investigated epicardial fat thickness measured by echocardiography and its relation to coronary artery calcification where they found that epicardial fat thickness has a statistically significant correlation to hypertension, diabetes mellitus, and LDL-C [15].

Also, the EAT mean thickness was found to be statistically significantly higher among those with significant coronary stenosis than those without [7.859 ± 0.691 mm versus 5.600 ± 0.386 mm] (P value < 0.001). This observation is supported by the study of Shambu et al. who measured EAT in 503 patients undergoing coronary angiogram and found that mean EAT was significantly higher in the CAD group than the control group (5.55 ± 1.21 mm vs 3.25 ± 1.15 mm, p < 0.0001) [14].

The results of our study are also concordant with Jeong et al. in 2007 who found that the EAT layer is significantly thicker in patients presenting with acute coronary syndrome compared to those having chronic stable angina and considered EAT as one of the risk factors of clinically significant CAD, along with other risk factors as age, diabetes mellitus, and smoking [16].

Moreover, on applying ROC curve analysis, the best discrimination between significant and non-significant CAD in relation to EAT thickness was found at cutoff value of > 6.1 mm, sensitivity was 100, specificity was 100, PPV was 100, NPV was 100 and accuracy was 100%. This result was supported by Eroglu et al. in 2009 who reported that the best cutoff value of EAT for CAD was 5.2 mm [5]. Also, Shambu et al. found that EAT at a cutoff ≥ 4.75 mm had 87% sensitivity and 63% specificity for the prediction of significant CAD [14].

All these findings could be attributed to and supported by Mazurek and collaborators who compared the expression of inflammatory factors in biopsy specimens of subcutaneous adipose tissue (SAT) and EAT in patients with severe coronary artery disease who did not have diabetes or obesity. The local expression of certain chemokines (MCP-1) and inflammatory cytokines (IL-1β, IL-6, and TNFα) was higher in EAT than in SAT, showing that EAT may contribute to the coronary local inflammatory potential. The author also observed a higher infiltration of macrophages in EAT than in SAT [17].

Relation between EAT and NAFLD

The current study showed a statistically significant relation between EAT thickness and degree of steatosis by ultrasound with higher values of EAT thickness measured among grade 3 hepatic steatosis than in grade 2 (9.600 ± 0.141 vs 8.256 ± 0.418 mm) (P value < 0.001). These results were in agreement with Lacobellis et al. who conclude that epicardial fat thickness was significantly higher (P < 0.01) in obese subjects with NAFLD when compared to those without NAFLD as well as in subjects with severe (ultrasound score 3) than those with moderate (score 2) liver steatosis (9.7 ± 0.2 vs. 8 ± 0.7 mm, P < 0.01) [18]. Also, the results of our study come in concordance with the findings of Stramaglia et al. in 2010 who reported that EAT measured by echocardiography was related to the severity of hepatic steatosis as detected by ultrasonography [19].

Also, a significant positive correlation was noted between EAT thickness and both ALP and GGT levels. These results are matching with Iacobellis et al. in 2008 who found a statistically significant relation between EAT on one side and ALT and AST on the other side [20].

Hence, our data suggested that echocardiography-measured epicardial fat thickness is independently associated with NAFLD. Its measurement can be clinically feasible and represent an additional tool for the stratification of the cardiometabolic risk in coronary heart disease.

Conclusions

High EAT thickness may represent a marker of the severity of NAFLD as well as an independent predictor of coronary artery disease risk. EAT assessment by trans-thoracic echocardiography is a simple and non-invasive tool that can be used for cardiovascular risk stratification.

Limitation of the study

The study includes rather a small number of cases and represents a single-center study. More advanced and expanded studies are recommended on a larger number of patients to get a better evaluation and to clarify the pathogenesis of the local fat tissues’ influence on systemic atherosclerosis.