Genes versus environment. The relationship between dietary fat and total and central abdominal fat

The influence of diet on body fat has not been quantified independently of genetic influences, although both are held to contribute to regulation of body fat stores. This study examined 1) the relationship between recent diet and total body and central abdominal fat in middle-aged female twins independent of genetic and important environmental factors and 2) evidence of interaction between diet and genetic predisposition.

Measurements in 436 healthy female twins (aged 58 +/- 10 years) included dietary intake by food frequency questionnaire (validated against a 7-day food diary, n = 162), BMI, total body and central abdominal fat by dual-energy X-ray absorptiometry, and environmental covariates (smoking habit, hormone replacement, and physical activity) by standardized questionnaire. Dietary energy underreporters were excluded.

Intake of dietary fat (total and subtype) and carbohydrates was not related to BMI or to total or central fat, confirmed in quintile analysis. With genetic and environmental factors controlled in 90 monozygotic pairs, differences in the intake of energy, fat, or protein were not related to intrapair differences in total and central body fat. However, a minor inverse relationship between carbohydrate intake and total adiposity was confirmed (r = -0.25, P = 0.02). In paired analyses, the twin with the higher intake of total sugars had significantly lower total body and central abdominal adiposity. There was no evidence of a gene-environment interaction between intake of fat or carbohydrates contributing to greater body fat mass in subjects genetically predisposed to obesity.

Using validated dietary measures and direct measures of body fat and excluding underreporters, no relationship between dietary fat and body fat was found in middle-aged women, particularly after controlling for genetic and some environmental factors. The role of dietary factors in determining total body and central abdominal fat appears to have been overestimated in past cross-sectional studies.