Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Randomized Controlled Trial
. 2017 Mar;105(3):589-599.
doi: 10.3945/ajcn.116.139683. Epub 2017 Feb 8.

Substituting Whole Grains for Refined Grains in a 6-wk Randomized Trial Favorably Affects Energy-Balance Metrics in Healthy Men and Postmenopausal Women

Affiliations
Free PMC article
Randomized Controlled Trial

Substituting Whole Grains for Refined Grains in a 6-wk Randomized Trial Favorably Affects Energy-Balance Metrics in Healthy Men and Postmenopausal Women

J Philip Karl et al. Am J Clin Nutr. .
Free PMC article

Erratum in

Abstract

Background: The effect of whole grains on the regulation of energy balance remains controversial.Objective: We aimed to determine the effects of substituting whole grains for refined grains, independent of body weight changes, on energy-metabolism metrics and glycemic control.Design: The study was a randomized, controlled, parallel-arm controlled-feeding trial that was conducted in 81 men and postmenopausal women [49 men and 32 women; age range: 40-65 y; body mass index (in kg/m2): <35.0]. After a 2-wk run-in period, participants were randomly assigned to consume 1 of 2 weight-maintenance diets for 6 wk. Diets differed in whole-grain and fiber contents [mean ± SDs: whole grain-rich diet: 207 ± 39 g whole grains plus 40 ± 5 g dietary fiber/d; refined grain-based diet: 0 g whole grains plus 21 ± 3 g dietary fiber/d] but were otherwise similar. Energy metabolism and body-composition metrics, appetite, markers of glycemic control, and gut microbiota were measured at 2 and 8 wk.Results: By design, body weight was maintained in both groups. Plasma alkylresorcinols, which are biomarkers of whole-grain intake, increased in the whole grain-rich diet group (WG) but not in the refined grain-based diet group (RG) (P-diet-by-time interaction < 0.0001). Beta ± SE changes (ΔWG compared with ΔRG) in the resting metabolic rate (RMR) (43 ± 25 kcal/d; P = 0.04), stool weight (76 ± 12 g/d; P < 0.0001), and stool energy content (57 ± 17 kcal/d; P = 0.003), but not in stool energy density, were higher in the WG. When combined, the favorable energetic effects in the WG translated into a 92-kcal/d (95% CI: 28, 156-kcal/d) higher net daily energy loss compared with that of the RG (P = 0.005). Prospective consumption (P = 0.07) and glycemia after an oral-glucose-tolerance test (P = 0.10) trended toward being lower in the WG than in the RG. When nonadherent participants were excluded, between-group differences in stool energy content and glucose tolerance increased, and between-group differences in the RMR and prospective consumption were not statistically significant.Conclusion: These findings suggest positive effects of whole grains on the RMR and stool energy excretion that favorably influence energy balance and may help explain epidemiologic associations between whole-grain consumption and reduced body weight and adiposity. This trial was registered at clinicaltrials.gov as NCT01902394.

Keywords: body fat; body weight; continuous glucose monitoring; energy metabolism; fiber; glucose tolerance; glycemia; hunger; resting metabolic rate; thermogenesis.

Figures

FIGURE 1
FIGURE 1
Geometric mean + SEM plasma total alkylresorcinol concentrations before and after the consumption of a refined grain–based diet or a whole grain–rich diet for 6 wk. *Significantly different from the refined-grain group at week 8 (P < 0.0001) and from the whole-grain group at week 2 (P < 0.0001) (repeated-measures ANOVA) (P-time-by-diet interaction < 0.0001).
FIGURE 2
FIGURE 2
Mean ± SEM combined changes in the RMR and fiber-adjusted stool energy content while consuming a refined grain–based diet or a whole grain–rich diet for 6 wk. *Significantly different from the refined-grain group, P = 0.005 (independent-samples t test). RMR, resting metabolic rate
FIGURE 3
FIGURE 3
Mean ± SEM changes in the glucose AUC (P = 0.10) and insulin AUC (P = 0.54) that were measured during an oral-glucose-tolerance test that was administered before and after the consumption of a refined grain–based diet (n = 13) or a whole grain–rich diet (n = 8) for 6 wk (A). Change scores were compared with the use of an ANCOVA with the preintervention value, age, BMI, sex, and the weight change during the 6-wk intervention entered as covariates. (B) Changes in total plasma alkylresorcinol concentrations were correlated with changes in glucose AUC (Pearson correlation). Data are presented with outliers included. The exclusion of outliers did not change the significance of any test. ARpost, alkylresorcinol measured after intervention; ARpre, alkylresorcinol measured before intervention.
FIGURE 4
FIGURE 4
Mean ± SEM weekly self-reported prospective consumption ratings during consumption of a 2-wk run-in diet and while consuming a refined grain–based diet (n = 40) or a whole grain–rich diet (n = 41) for 6 wk. Main and interactive effects of time and diet were examined with the use of marginal models with an unstructured covariance structure. Covariates in the model were means of ratings measured during the run-in phase (weeks 1 and 2), age, BMI, sex, and the weight change during the 6-wk intervention.

Comment in

Similar articles

See all similar articles

Cited by 18 articles

See all "Cited by" articles

Publication types

Associated data

Feedback