Characteristics of the included studies
This study followed the protocols specified in the PRISMA statement, and PRISMA checklist was shown in supplement 1 file. A total of 794 studies initially identified after searching the database of PubMed, Embase, and the Cochrane. 47 observational studies matched our eligibility criteria and 39 remained after exclusion for reasons: non-English(n = 1), reported the studies from the same original database (n = 2), secondary hypertension (n = 2), comprised insufficient data (n = 3). A flowchart of the study selection was shown in Fig. 1. The Newcastle–Ottawa scale (NOS) was used to evaluate the quality of the articles, and these studies were assessed as high-quality publications (median: 8; 25th and 75th percentile: 7 and 9). Accordingly, A total of 8,983 subjects were included in this study. There were 18 studies comparing EAT thickness and/or volume between EH patients and normotensive subjects11–14,16−18,21,23−32, including 13 studies reported EAT-t, 3 studies reported EAT-v, and 2 articles reported both EAT-t and EAT-v between the two groups. There were 16 studies dividing the subjects (including hypertensive and normotensive subjects) into two groups according to EAT thickness or volume14,33−47. Seven studies compared the difference of EAT between non-dipper and dipper hypertension13,18,27,32,48−50. Two studies investigated the association of EAT-t with the grade of hypertension12,30. Additionally, four studies reported the association of cardiac remodeling and function with EAT15,26,51,52,and three studies provided data addressing the association of ASCVD with EAT in hypertensive patients15,28,53.
The association of epicardial adipose tissue thickness and volume with essential hypertension
There were 18 studies compared the EAT-t and EAT-v between 3258 hypertensive patients and 1928 normotensive controls. We performed a meta-analysis of comparing EAT-t and EAT-v between hypertensive and normotensive, and the results showed that both EAT-t (EAT-t: SMD = 0.64, 95% CI: 0.44–0.83, p < 0.001; Fig. 2A) and EAT-v (EAT-v: SMD: 0.69, 95% CI:0.34 − 0.1.05, p < 0.001; Fig. 2A) were significantly increased in hypertensive compared to those in normotensive. TSA showed that the cumulative z-curve crossed the boundary for futility (TSA adjusted 95% CI: 0.72–1.55; p < 0.001; Fig. 2B) and a definite conclusion can be drawn. A total of 4795 subjects (normotensive subjects and hypertensive patients) were divided into two groups according to EAT thickness or volume in sixteen studies, and we investigated the association of EAT-t and EAT-v with hypertension. The meta-analysis indicated that EAT-t (OR: 1.59, 95% CI: 1.09–2.33, P < 0.001; Fig. 3A) and EAT-v (OR: 1.82, 95% CI: 1.33–2.19, P < 0.001; Fig. 3A) were strongly associated with the occurrence of EH. The TSA confirmed that the results were reliable (TSA-adjusted 95% CI: 1.31–2.21; p < 0.001; Fig. 3B).
Comparison of the epicardial adipose tissue thickness and volume between dipper and non-dipper hypertension
Six studies compared the EAT-t and one study compared the EAT-v between dipper and non-dipper hypertension. We performed a meta-analysis to compare EAT-t and EAT-v between dipper and non-dipper hypertension. The results showed that both EAT-t (SMD = 0.85, 95% CI = 0.49 − 0.1.21, p < 0.001; Fig. 4) and EAT-v (SMD = 0.83, 95% CI = 0.31–1.34, p = 0.002; Fig. 4) were increased in non-dipper hypertensive patients than those in dipper hypertensive patients.
Comparison of the epicardial adipose tissue thickness among the grades of hypertension
Two studies compared the epicardial adipose tissue thickness among the patients with different grades of hypertension. We performed a meta-analysis to compare EAT-t between grade 1 and grade 2 hypertension and between grade 2 and grade 3 hypertension. However, there was no significant differences in EAT-t among the patients with different grades of essential hypertension (grade 1 to grade 2: SMD = 0.23, 95% CI=-0.13-0.59, p = 0.211; grade 2 to grade 3: SMD = 0.09, 95% CI=-0.15-0.33, p = 0.452; Fig. 5).
The association of epicardial adipose tissue with arteriosclerotic cardiovascular disease and cardiac morphology and function in hypertensive patients
We also investigate the association of EAT with the complication in EH. We compare the EAT-t in EH patients with normal and abnormal cardiac morphology. Higher mean of EAT-t (SMD = 0.85, 95% CI = 0.71–0.99, p < 0.001; Fig. 6A) was found in patients with cardiac hypertrophy and dysfunction than that in patients with normal cardiac morphology and function. Additionally, we also compared the EAT in EH patients with and without arteriosclerotic cardiovascular disease (ASCVD), and the results showed that EAT-t (SMD = 0.37, 95% CI = 0.23–0.52, p < 0.025; Fig. 6B) was increased in hypertensive patients with ASCVD than those without.
There were significant differences in EAT thickness and volume mean values among individual studies (I2 = 86%, p < 0.001). Similarly, we found significant differences in the association of EAT with essential hypertension among individual studies (I2 = 70.4%, p < 0.001). Therefore, a random-effects model was used to analyze the data. Publication bias about the mean of EAT between essential hypertension and controls were assessed and shown in Fig. 7A. The publication bias reporting the association of EAT with essential hypertension were determined and shown in Fig. 7B. The results showed that there was no significant publication bias.