Visceral and Intrahepatic Fat Are Associated with Cardiometabolic Risk Factors Above Other Ectopic Fat Depots: The Framingham Heart Study

Abstract

Background: We examined the associations among 8 different fat depots accumulated in various anatomic regions and the relationship between these fat depots and multiple cardiometabolic risk factors.

Methods: Participants were from the Framingham Heart Study Offspring and Third Generation who also participated in the multidetector computed tomography substudy in 2002-2005. Exposures were multidetector computed tomography-derived fat depots, including abdominal subcutaneous adipose tissue, abdominal visceral adipose tissue, intramuscular fat, intrathoracic fat, pericardial fat, thoracic periaortic fat, intrahepatic fat, and renal sinus fat. Multivariable-adjusted regression analyses with a forward selection procedure were performed to identify the most predictive fat depots.

Results: Of 2529 participants, 51.9% were women (mean age, 51.1 years). Visceral adipose tissue had the strongest correlations with each of the other fat measures (range, 0.26-0.77) and with various cardiometabolic risk factors (range, -0.34 to 0.39). As determined by the selection models, visceral adipose tissue was the only fat depot that was associated with all cardiometabolic risk factors evaluated in this study (all P<.05). Selection models also showed that subcutaneous adipose tissue and intrahepatic fat were associated with cardiometabolic risk factors related to the traits of dysglycemia, dyslipidemia, and hypertension (all P<.05). However, only associations with visceral adipose tissue and intrahepatic fat persisted after further adjustment for body mass index and waist circumference.

Conclusions: Visceral adipose tissue and intrahepatic fat were consistent correlates of cardiometabolic risk factors, above and beyond standard anthropometric indices. Our data provide important insights for understanding the associations between variations in fat distribution and cardiometabolic abnormalities.

Keywords: Cardiometabolic risk factors; Computed tomography; Ectopic fat; Epidemiology.

Figure 1

Heat map of the association a) among fat measures; b) between fat measures and cardiometabolic risk factors based on age and sex-adjusted partial Pearson correlation coefficients (r). The color depth of each cell represents the magnitude of the absolute values of the correlation coefficients from low (|r|=0 for Figure 1a and 1b; yellow) to high (|r|=1 for Figure 1a and |r|=0.5 for Figure 1b; red). Unless otherwise specified, all results were significant at p<0.05. ^ap>0.05. For Figure 1b, correlation coefficients associated with higher abdominal subcutaneous adipose tissue and visceral adipose tissue, intrathoracic fat, pericardial fat, thoracic periaortic fat, and renal sinus fat; and lower intramuscular fat and intrahepatic fat are presented. Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein; MDCT, multi-detector computed tomography.

Figure 1

Heat map of the association a) among fat measures; b) between fat measures and cardiometabolic risk factors based on age and sex-adjusted partial Pearson correlation coefficients (r). The color depth of each cell represents the magnitude of the absolute values of the correlation coefficients from low (|r|=0 for Figure 1a and 1b; yellow) to high (|r|=1 for Figure 1a and |r|=0.5 for Figure 1b; red). Unless otherwise specified, all results were significant at p<0.05. ^ap>0.05. For Figure 1b, correlation coefficients associated with higher abdominal subcutaneous adipose tissue and visceral adipose tissue, intrathoracic fat, pericardial fat, thoracic periaortic fat, and renal sinus fat; and lower intramuscular fat and intrahepatic fat are presented. Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein; MDCT, multi-detector computed tomography.