Effects of matching diet type to obesity-related genotype on body composition changes in women during a six-month resistance-exercise training and walking program

Abstract

Methods Fifty sedentary, obese women (41.6 ± 12 yrs, 35.4 ± 8 kg/ m) were assigned to diet groups based on five obesityrelated genetic variants from four genes prominently associated with obesity (FABP2, PPARG, ADRB2, ADRB3). Participants were either truly matched (T) to their diet group based on genotype (n = 28) or falsely matched (F) based on genotype (n = 22). Prescribed diets consisted of 1,500 kcal/d and included carbohydrate:fat: protein percentages of 30:25:45 (H) or 20:35:45 (L). Participants performed a supervised circuit-style resistanceexercise program four days/week and a walking program consisting of 10,000 steps/day, three days/week. Body weight and duel energy X-ray absorptiometry (DXA) body composition measures were obtained at baseline, 4, 8, 12, 16, 20, and 24 weeks. Data were analyzed by MANOVA, with baseline body weight and body composition values used as covariates to normalize baseline differences between groups. Data are presented as changes from baseline at each time point, respectively. Results MANOVA revealed an overall Wilks’ Lamda time effect (p < 0.001) with no significant time by diet (p = 0.51), time × gene type (0.84), or time × diet × gene type (0.81) effects observed. Univariate analysis revealed that the exercise and diet interventions promoted significant reductions in weight (-5.36 ± 5.0 kg, p < 0.001), fat mass (-4.53 ± 3.6 kg, p < 0.001), and body fat (-2.88 ± 2.7 %, p < 0.001) with a trend toward a reduction in fat free mass (-0.65 ± 2.3 kg, p < 0.071). When baseline body weight and DXA body composition variables were used as covariates, Wilks’ Lambda time × diet (p = 0.098) tended to differ, a time × gene type interaction was observed (p = 0.011), while no differences were seen in time × diet × gene type (p = 0.18). Univariate analyses revealed some trends in time × diet changes in weight (H -2.03 ± 1.7, -3.13 ± 2.6, -4.17 ± 3.3, -4.62 ± 4.0, -4.75 ± 4.6, -4.41 ± 5.1; L -2.47 ± 1.8, -3.66 ± 2.4, -4.56 ± 3.1, -5.49 ± 3.9, -5.89 ± 4.5, -6.17 ± 4.8 kg, pq = 0.02), fat mass (H -1.53 ± 1.4, -2.67 ± 2.3, -3.63 ± 2.5, -3.73 ± 2.8, -4.14 ± 3.6, -3.95 ± 3.6; L -1.31 ± 1.6, -2.66 ± 2.2, -3.22 ± 2.5, -4.32 ± 2.9, -4.60 ± 3.0, -5.03 ± 3.7 kg, pq = 0.10), FFM (H -0.49 ± 1.2, -0.58 ± 1.5, -0.52 ± 1.8, -0.68 ± 2.4, -0.51 ± 2.2, -0.31 ± 2.3; L -0.95 ± 1.5, -0.73 ± 1.9, -1.20 ± 1.9, -0.86 ± 2.0, -1.03 ± 2.6, -0.94 ± 2.4 kg, p = 0.14), or body fat (H -0.50 ± 1.9, -1.51 ± 1.8, -2.32 ± 2.2, -2.22 ± 2.0, -2.68 ± 2.5, -2.65 ± 2.3; L -0.46 ± 1.5, -1.55 ± 2.1, -1.67 ± 2.4, -2.56 ± 2.5, -2.78 ± 2.7, -3.08 ± 3.0 %, pq = 0.13) generally in favor of the more carbohydrate restricted diet. Some trends were also seen in time × gene type changes in weight (T -2.06 ± 1.8, -2.91 ± 2.6, -3.99 ± 3.3, -4.83 ± 4.0, -5.07 ± 4.6, -5.15 ± 5.1; F -2.54 ± 1.7, -4.05 ± 2.2, -4.88 ± 2.9, -5.43 ± 3.8, -5.73 ± 4.6, -5.62 ± 5.0 kg, p = 0.20), Exercise & Sport Nutrition Lab, Texas A&M University, College Station, TX, USA Full list of author information is available at the end of the article Coletta et al. Journal of the International Society of Sports Nutrition 2015, 12(Suppl 1):P16 http://www.jissn.com/content/12/S1/P16