Response to “Epicardial and Pericardial Fat: Close, but Very Different”

TO THE EDITOR: First we thank Icobellis for his interest in our work and more importantly express the value of his contributions related to obesity and heart disease, which have been pioneering, and influential to us and other groups. We appreciate many of the comments in the letter titled “Epicardial and Pericardial Fat: Close, but Very Different” but respectively disagree with others. Importantly, we agree about the need to use scientific language precisely and that the pericardial anatomy is especially confusing. The term “Pericardial” fat includes both epicardial and paracardial adipose tissues. Icobellis states, “Anatomically, epicardial and pericardial adipose tissue are clearly different” and references the recent article by Sacks (1). In this article, “pericardial fat” is described, consistent with our definition, as encompassing both epicardial fat (located within or deep into the pericardium) and paracardial fat (located superficial to the pericardium). Directly quoting Sacks, “Pericardial fat (pericardial adipose tissue) is defined as epicardial fat in all these possible locations plus paracardial fat.” (1) Likewise, Taguchi et al. measured the volume of pericardial fat using a comparable method and definition of the anatomical spaces (2). Thus, our use of “pericardial fat” is consistent with the current anatomical usage and encompasses both epicardial and paracardial locations. We completely agree that there are numerous established differences between adipose tissue located in the paracardial space and the epicardial space. The key point, based on the biology and anatomic spaces, is that to distinguish true epicardial tissue from paracardial adipose tissue one must identify the pericardium (Figure 1). The normal pericardium is commonly difficult to be visualized by all imaging modalities including echocardiography, even over the right ventricle free wall (3). Small pericardial effusions can be missed secondary to a variety of reasons. We do not believe that one can assume that all echogenic material adjacent to the free wall of the right ventricle is epicardial fat. As demonstrated in Figure 1 both paracardial and epicardial fat surround the right ventricle. Magnetic resonance imaging (MRI) indirectly images the “pericardial complex” secondary to spin de-phasing of pericardial fluid. Modern cardiac computed tomography (CT) does have the spatial and temporal resolution to image the pericardium directly. Current cardiac MRI and CT technology can differentiate the epicardial space from the paracardial space. Note that this task becomes increasingly difficult in lean individuals who have limited pericardial adipose tissues. This is a single image from cardiac computed tomography angiography exam using submillimeter spatial resolution and cardiac electrocardiogram gating. The pericardium over the free wall of the right ventricle can be clearly delineated and measurement of the thickness of the epicardial fat (D = 2.8 mm) and paracardial fat (E = 7.2 mm) is clearly demonstrated. If one fails to clearly identify the pericardium, misclassification of the amount of epicardial and paracardial fat will occur. Why did we choose to measure “pericardial fat” rather than “epicardial” fat? Using a dedicated approach we have measured epicardial adipose tissue in a substudy of 159 participants. The Spearman's correlation coefficient was 0.92 (P < 0.0001) indicating a very high association. Thus people with large amounts of epicardial fat have large amounts of pericardial fat (i.e., paracardial + epicardial adipose tissue). Our research question is to understand better the role of adiposity and the heart in lean, average, and obese individuals who are asymptomatic. Thus, we wanted to make sure that we did not create a bias related to measurement in those lean individuals in whom measuring epicardial fat is very difficult and in whom misclassification of paracardial fat as epicardial fat would result in a significant measurement error. Likewise, both precision and reproducibility of the measurement is important for population-based and genomic studies where power to detect changes and adjust for multiple comparisons is critically important. Lastly, logistically we need a protocol that could be accomplished in over 10,000 cardiac CT exams. Measuring pericardial adipose tissue (i.e., paracardial and epicardial) reduces or eliminates errors related to misclassification of the two adjacent compartments when the pericardium is not clearly delineated and provides a time-efficient protocol for this initial phase of scientific inquiry. We, like others, believe that understanding all the adipose tissue in and around the heart will be very informative: intramyocardial, epicardial, and paracardial. Our approach is a step in better understanding this exciting and important area of science. The authors declared no conflict of interest.