Screening and selection of human volunteers at increased cardio-metabolic risk for a dietary intervention study, with particular reference to levels of liver fat

Abstract

Background: Individuals at increased cardio-metabolic (CM) risk have a greater chance of developing diabetes mellitus and suffering premature cardiovascular events. The metabolic syndrome (MetS) is a constellation of risk factors which confer increased CM risk. MetS was originally proposed by Reaven (1998) and the currently accepted definition was agreed by AHA/NHLBI,WH,IASIASO (Alberti et al., 2009). More recently, increased liver fat has been proposed as a marker of CM risk. The aims of this study were to evaluate the inter-relationships between percentage liver fat and established CM risk factors and identify those with MetS. Methods: This paper presents a secondary analysis of baseline data from a sample (n = 37) (56 ± 5.32 years), of overweight men who were at increased CM risk but otherwise healthy,, and recruited into a randomised dietary intervention crossover study to investigate how dietary carbohydrate influences the metabolism of plasma lipoproteins in overweight volunteers with moderately high (>10% <20%) and low (<2%) levels of liver fat. Anthropometric and biochemistry measurements were obtained, and from a sub-sample, the percentage of liver fat by magnetic resonance imaging. The participants were also assigned a metabolic score based on the number of indicators of the MetS according to the AHA/NHLBI, WH, IASIASO definition (Alberti, 2009). Non-parametric statistical analyses was used, as not all variables were normally distributed, there were a number of outliers and a small sample size. Pearson Correlation Co-efficients were used to investigate relationships between continuous and categorical variables. The variables were not adjusted for age or BMI because of a small sample size. Statistical significance was set at P ≤ 0.05 and all analyses were carried out using SPSS 16.0. Results: Fifty nine per cent of the participants had MetS (n = 21). There was evidence of significant, positive associations between the metabolic score for MetS and total cholesterol:HDL-C ratio (P = 0.030), and between percentage liver fat and plasma glucose and insulin (P = 0.010 and P = 0.003 respectively). The percentage of liver fat was also notably higher than that reported in other studies (Yamada et al., 2010; Kotronen et al., 2007) and there was positive correlation between the liver enzyme alanine transaminase (ALT) and plasma insulin (P = 0.012). Conclusion: The strong associations between established risk factors and liver fat suggest that the latter may be clinically useful as a marker for CM risk in MetS.