4.1 Main results
This study mainly showed an association between increasing EAT and AFACS occurrence. To investigate the relationship between EAT volume, both total and LA specified EAT, as well as EAT thickness and AFACS occurrence, three different meta-analyses were conducted. The findings were similar, which showed a positive association between increased EAT and AFACS occurrence. Surprisingly, the similar findings in EAT thickness and EAT volume put emphasis on the value of echocardiography in evaluation of EAT, which is a safe and non-invasive imaging modality24.This study, the first attempt to prove the association between EAT and AFACS occurrence, suggested a new possible imaging marker to predict AFACS, which will require further research.
4.2 EAT and AFACS
Although the mechanism of increased EAT leading to AFACS is still unclear, studies that demonstrated association between abundant EAT and AF can be taken into account. EAT locates between the myocardium and visceral pericardium, mainly on the right ventricle surface and anterior wall of the left ventricle. It also surrounds atrioventricular grooves and great coronary vessels, reaching the main thickness at the anterior and lateral walls of right atrium 25. Due to the position, It exerts local paracrine effect with findings of increased expression of numerous inflammatory markers, and as a metabolically active tissue modulating the adjacent myocardial tissue with abundant ganglionated plexus of autonomic nervous system15,26,27. At the same time, abundant EAT has been illustrated to be associated with fatty infiltration into adjacent myocardium by histopathological evidence 28. Mahajan and Zghaib et al 29,30 proved EAT could cause lower bipolar voltage and electrogram fractionation as electrophysiologic substrates for AF. Besides, there is a speculate that the imbalance between excessive oxidative stress and the protective effect of adiponectin ultimately leads to the occurrence of AF31.
The similarity is that AFACS associates with a combination of factors: clinical variables, intraoperative surgical variables, electrocardiograph markers and echocardiographic predictors. In detail, advanced age, hypertension, male gender, right coronary artery stenosis, depressed left ventricular function, a remote history of previous AF, combined valve replacement/coronary artery bypass grafting procedures, prolonged aortic cross-clamp and bypass times, PR-interval, QRS duration, abnormal left ventricular systolic and diastolic function, left ventricular hypertrophy and increased left atrial volume may predispose to postoperative AF 32-36. And AFACS was also proved to be associated with the occurrence of systemic inflammatory response syndrome with inflammatory markers, including IL-6, IL-8, C-reactive protein, tumor necrosis factor-α and indices of neutrophil and platelet activation, increasing significantly in the systemic bloodstream 37,38. Electrolyte imbalance, especially hypokalaemia, is also considered to be a triggering factor of AFACS 37,39.
It has been proved that obesity or a high BMI was a risk factor for AF40,41. Compared with non-obese people, obese individuals have a higher probability (49%) of suffering with AF, which may be related with atrial enlargement, ventricular diastolic dysfunction and increased ectopic fat depot 41. And, in other reports, EAT was proved to be associated with AF even after adjustment for AF risk factors, including a high BMI13,42 . However, Echahidi et al. 37reported that metabolic syndrome and obesity were independent risk factors for AFACS. And EAT is associated with obesity or high BMI42. So, whether EAT should be regarded as an independent trigger or a modulator of obesity of AFACS requires further delineation.
In this study, it is known that increased EAT is associated with AFACS, soreducing EAT maybe potential to decrease the AFACS. Antidiabetic drugs were used for diabetes mellitus patients, which might decrease EAT and the accumulation of EAT, including Thiazolidinediones43, glucagon-like peptide 1 receptor agonists44, Dipeptidyl peptidase-4 inhibitors45 and Sodium-dependent glucose transporters 2 inhibitors46. Besides, weight loss by diet control or exercise suggested to decrease EAT in interventricular groove, which suggests a correlation with improved insulin resistance and insulin sensitivity indexes47. A control study (in vivo and in vitro) performed on aortic stenosis patients who underwent a cardiac surgery suggested that statin therapy was related to EAT thickness reduction, which seemed secondary to the characteristics of statins anti-inflammatory effect48. Considering EAT containing large amount of ganglionated plexi, an increasing number of studies aimed at lowered AF inducibility due to injecting botulinum toxin into canine epicardial autonomic ganglia49. Although it was hypothetical, still promising to be a potential novel therapeutic option. All of these therapies not only might reduce EAT, but decrease incidence of many other diseases, especially metabolic diseases which is closely associated with AFACS.
Ten studies fell into sub-groups as per measuring methods of EAT, including echocardiography and CT. Comparatively, the results measured by CT were more preferred than echocardiography, because EAT was not uniformly distributed around the heart. Echocardiographic probe angulation on 2-dimensional imaging could decrease reproducibility and accuracy, as well as unable to measure periatrial fat or total EAT volume. And echocardiographic measurement might increase the EAT thickness, because overlapping with perivascular adipose tissue. But when patients had coronary atherosclerosis history, which enlarged thickness of EAT, image analysis probably provided accurate measurement. Accordingly, Nerlekar et.al50 proved that EAT measured by CT compared with echocardiography revealed poor correlation, and discrepancy was particularly obvious at higher EAT thickness.