A randomized controlled trial: the effect of inulin on weight management and ectopic fat in subjects with prediabetes

Abstract

Background: Fat infiltration of the liver, muscle and pancreas is associated with insulin resistance and risk of diabetes. Weight loss reduces ectopic fat deposition and risk of diabetes, but is difficult to sustain to due to compensatory increases in appetite. Fermentable carbohydrates have been shown to decrease appetite and food intake, and promote weight loss in overweight subjects. In animal studies, fermentable carbohydrate reduces ectopic fat independent of weight loss. We aimed to investigate the effect of the fermentable carbohydrate inulin on weight maintenance, appetite and ectopic fat in subjects with prediabetes.

Methods: Forty-four subjects with prediabetes were randomized to 18 weeks' inulin or cellulose supplementation. During weeks 1-9 (weight loss phase) all subjects had four visits with a dietitian to guide them towards a 5 % weight loss. During weeks 10-18 (weight maintenance phase) subjects continued taking their assigned supplementation and were asked to maintain the weight they had lost but were offered no further support. All subjects attended study sessions at baseline, 9 and 18 weeks for measurement of weight; assessment of adipose tissue and ectopic fat content by magnetic resonance imaging and magnetic resonance spectroscopy; glucose, insulin and GLP-1 levels following a meal tolerance test; and appetite by ad libitum meal test and visual analogue scales.

Results: Both groups lost approximately 5 % of their body weight by week nine (-5.3 ± 0.1 % vs -4.3 ± 0.4 %, p = 0.13, but the inulin group lost significantly more weight between 9 and 18 weeks (-2.3 ± 0.5 % vs -0.6 ± 0.4 %, p = 0.012). Subjects taking inulin had lower hepatic (p = 0.02) and soleus muscle (p < 0.05) fat content at 18 weeks compared to control even after controlling for weight loss and consumed less at the ad libitum meal test (p = 0.027). Fasting glucose significantly decreased at week nine only (p = 0.005), insulin concentrations did not change, and there was a significant increase in GLP-1 in the cellulose group at 9 and 18 weeks (p < 0.03, p < 0.00001).

Conclusion: Inulin may have a two-pronged effect on the risk of diabetes by 1) promoting weight loss 2) reducing intrahepatocellular and intramyocellular lipid in people with prediabetes independent of weight loss.

Clinical trial number: NCT01841073.

Keywords: Appetite; Carbohydrate; Diabetes prevention; Diabetes risk; Fibre; Intrahepatocellular lipid; Intramyocellular lipid; Weight management.

Figures

Fig. 1

Schematic showing study outline, including the timings of blood samples, VAS and breath hydrogen measure during the MTT. H2: breath hydrogen measure; MTT: meal tolerance test; MRS: magnetic resonance imaging; MRS: magnetic resonance spectroscopy; VAS: visual analogue scales

Fig. 2

Percentage weight loss and body fat loss measured by BIA at week nine and week 18 in inulin and cellulose groups. Analysis was done by ANCOVA with baseline weight as a covariate. Weight loss at week nine was not significantly different between inulin and cellulose groups (−5.3 ± 0.1 %, n = 20 vs −4.3 ± 0.4 %, n = 19, p = 0.13). Between weeks 9–18 the inulin group lost significantly more than the cellulose group (−2.3 ± 0.5 %, n = 20 vs −0.6 ± 0.4 %, n = 18, p = 0.012). Analysis for body loss was done by ANCOVA with baseline weight as a covariate. The inulin group lost a greater percentage of body fat as measured by BIA at 9 (−2.8 ± 0.4 %, n = 20 vs −1.2 ± 0.4 %, n = 19, p < 0.01) and 18 weeks (−3.7 ± 0.6 %, n = 20 vs −1.1 ± 0.6 %, n = 18, p = 0.01). ANCOVA: analysis of covariance; BIA: bioelectrical impedance

Fig. 3

Change in intrahepatocellular lipid (IHCL) at weeks 9 and 18 in inulin and cellulose groups. Analysis was done by ANCOVA with change in body weight as a covariate. IHCL was significantly reduced in subjects randomised to the inulin supplement compared to the cellulose at 9 (−9.6 ± 2.8 %, n = 10 vs −0.5 ± 2.7 %, n = 9, p < 0.04) and 18 weeks (−10.0 ± 2.6 %, n = 9 vs −2.3 ± 2.5 %, n = 5, p = 0.02). ANCOVA: analysis of covariance; IHCL: intrahepatocellular lipid

Fig. 4

Change in intramyocellular lipid in the soleus muscle (IMCL-S) at weeks 9 and 18 in inulin and cellulose groups. Analysis was done by ANCOVA with change in body weight as a covariate. IMCL-S was significantly reduced at 9 and 18 weeks in the inulin group compared to cellulose: (9 weeks: −0.7 ± 0.3 %, n = 10 vs 0.8 ± 0.3 %, n = 9, p < 0.005); (18 weeks: −1.3 ± 1.4 %, n = 9 vs 4.8 ± 3.0 %, n = 5, p < 0.05). ANCOVA: analysis of covariance; IHCL: intrahepatocellular lipid

Fig. 5

Time course data for plasma glucose at baseline, 9-weeks and 18 weeks for inulin (i) and cellulose supplementation (ii). Black line = baseline; dark grey line = 9-week visit; light grey line = 18-week visit. tAUC was calculated using the trapezoid method. ANCOVA using weight change as a covariate was used to analyse delta change between the inulin and cellulose groups. Glucose tAUC delta change at week nine(135 ± 134, n = 17 vs 87 ± 42, n = 17, p = 0.37) or at week 18 (−267 ± 104, n = 17 vs −38 ± 21, n = 17, p = 0.37) did not differ between groups once controlled for weight loss. There was a significant difference in the FPG delta change between groups at week nine once controlled for weight loss (−0.23 ± 0.17 mmol/L, n = 20 vs 0.44 ± 0.24 mmol/L, n = 19, p = 0.005) while the change at week 18 was no longer significant (−0.40 ± 0.19 mmol/L, n = 20 vs 0.16 ± 0.23 mmol/L, n = 18, p = 0.08) (iii). ANCOVA: analysis of covariance; FPG; fasting plasma glucose; tAUC: total area under the curve

Fig. 6

Time course data for plasma insulin at baseline, 9 weeks and 18 weeks for inulin (i) and cellulose supplementation (ii). Black line = baseline; dark grey line = 9-week visit; light grey line. tAUC was calculated using the trapezoid method. ANCOVA using weight change as a covariate was used to analyse delta change between the inulin and cellulose groups. After controlling for weight lost the delta change in insulin tAUC following inulin supplementation was similar to the cellulose group at week nine (−2366 ± 575, n = 17 vs −1566 ± 1724, n = 17, p = 0.66) and week 18 (−2643 ± 671, n = 17 vs −1264 ± 1045, n = 17, p = 0.27). The delta change in fasting insulin was also similar between inulin and cellulose at week nine (−23.3 ± 15.4 pmol/L, n = 18 vs −30.5 ± 17.6 pmol/L, n = 17, p = 0.82) and week 18 (−35.5 ± 17.8 pmol/L, n = 18 vs −14.7 ± 18.7 pmol/L, n = 17, p = 0.53). ANCOVA: analysis of covariance; tAUC: total area under the curve