Associations of liver fat content with cardiometabolic phenotypes and outcomes in a multi-ethnic population: Results from the Dallas Heart Study.

Abstract

To evaluate the associations between liver fat content and cardiometabolic parameters to explore potential threshold values that define metabolically healthy liver fat content, and to examine the association of liver fat content with cardiovascular events as well as its longitudinal progression. Participants in the Dallas Heart Study underwent clinical evaluation, including laboratory testing, and liver fat quantification by magnetic resonance spectroscopy (MRS) at baseline (N=2287) and at follow-up (N=343) after a mean of 7.3 years. Cardiovascular events were adjudicated (>12 years). The mean age at study entry was 44 years, 47% of participants were men, and 48% were African American. The following cardiometabolic biomarkers worsened across liver fat quintiles (P < 0.0001): body mass index (BMI); waist circumference; prevalence of hypertension; prevalence of diabetes; cholesterol, triglyceride, high-sensitivity C-reactive protein (CRP), leptin and fasting glucose levels; homeostatic model assessment of insulin resistance index (HOMA-IR); coronary artery calcium score; visceral adipose tissue; abdominal subcutaneous adipose tissue; and lower body subcutaneous adipose tissue. Cardiovascular events were comparable across groups defined by tertile of baseline liver fat content. Change in BMI (R=0.40), waist circumference (R=0.35), CRP (R=0.31), alanine aminotransferase (R=0.27), HOMA-IR (R=0.26), aspartate transaminase (R=0.15) and triglycerides (R=0.12) significantly correlated with change in liver fat content (P < 0.01 for all). Clinically relevant metabolic abnormalities were higher across quintiles of liver fat, with increases noted well within normal liver fat ranges, but cardiovascular events were not associated with liver fat content. Longitudinal changes in metabolic parameters, especially adiposity-related parameters, were correlated with change in liver fat content.