Impact of dietary and obesity genetic risk scores on weight gain

Abstract

ABSTRACT Background Whether genetic background and/or dietary behaviors influence weight gain in middle-aged subjects is debated. Objective To assess whether genetic background and/or dietary behaviors are associated with changes in obesity markers (BMI, weight, and waist and hip circumferences) in a Swiss population-based cohort. Methods Cross-sectional and prospective (follow-up of 5.3 y) study. Two obesity genetic risk scores (GRS) based on 31 or 68 single nucleotide polymorphisms were used. Dietary intake was assessed using a semiquantitative FFQ. Three dietary patterns “Meat & fries” (unhealthy), “Fruits & vegetables” (healthy), and “Fatty & sugary” (unhealthy), and 3 dietary scores (2 Mediterranean and the Alternative Healthy Eating Index [AHEI]) were computed. Results On cross-sectional analysis (N = 3033, 53.2% females, 58.4 ± 10.6 y), obesity markers were positively associated with unhealthy dietary patterns and GRS, and negatively associated with healthy dietary scores and patterns. On prospective analysis (N = 2542, 54.7% females, age at baseline 58.0 ± 10.4 y), the AHEI and the “Fruits & vegetables” pattern were negatively associated with waist circumference gain: multivariate-adjusted average ± SE 0.96 ± 0.25 compared with 0.11 ± 0.26 cm (P for trend 0.044), and 1.14 ± 0.26 compared with –0.05 ± 0.26 cm (P for trend 0.042) for the first and fourth quartiles of the AHEI and the “Fruits & vegetables” pattern, respectively. Similar inverse associations were obtained for changes in waist >5 cm: multivariate-adjusted OR (95% CI): 0.65 (0.50, 0.85) and 0.67 (0.51, 0.89) for the fourth versus the first quartile of the AHEI and the “Fruits & vegetables” dietary pattern, respectively. No associations were found between GRS and changes in obesity markers, and no significant gene-diet interactions were found. Conclusion Dietary intake, not GRS, are associated with waist circumference in middle-aged subjects living in Lausanne, Switzerland.