Drinks containing anthocyanin-rich blackcurrant extract decrease postprandial blood glucose, insulin and incretin concentrations

Abstract

Blackcurrants are rich in polyphenolic glycosides called anthocyanins, which may inhibit postprandial glycemia. The aim was to determine the dose-dependent effects of blackcurrant extract on postprandial glycemia. Men and postmenopausal women (14M, 9W, mean age 46 years, S.D.=14) were enrolled into a randomized, double-blind, crossover trial. Low sugar fruit drinks containing blackcurrant extract providing 150-mg (L-BE), 300-mg (M-BE) and 600-mg (H-BE) total anthocyanins or no blackcurrant extract (CON) were administered immediately before a high-carbohydrate meal. Plasma glucose, insulin and incretins (GIP and GLP-1) were measured 0-120min, and plasma 8-isoprostane F2α, together with arterial stiffness by digital volume pulse (DVP) was measured at 0 and 120min. Early plasma glucose response was significantly reduced following H-BE (n=22), relative to CON, with a mean difference (95% CI) in area over baseline (AOB) 0-30min of -0.34mmol/l.h (-0.56, -0.11, P<.005); there were no differences between the intermediate doses and placebo. Plasma insulin concentrations (AOB 0-30min) were similarly reduced. Plasma GIP concentrations (AOB 0-120min) were significantly reduced following H-BE, with a mean difference of -46.6ng/l.h (-66.7, -26.5, P<.0001) compared to CON. Plasma GLP-1 concentrations were reduced following H-BE at 90min. There were no effects on 8-isoprostane F2α or vascular function. Consumption of blackcurrant extract in amounts roughly equivalent to 100-g blackcurrants reduced postprandial glycemia, insulinemia and incretin secretion, which suggests that inclusion of blackcurrant polyphenols in foods may provide cardio-metabolic health benefits. This trial was registered at clinicaltrials.gov as NCT01706653.

Keywords: Anthocyanins; Berries; Incretins; Insulin; Polyphenols; Postprandial glycemia; Randomized controlled trial.

Copyright © 2016 The Authors. Published by Elsevier Inc. All rights reserved.

Figures

Fig. 1

Study visit protocol.

Fig. 2

Consort diagram.

Fig. 3

Postprandial plasma glucose and insulin concentrations. Mean (±S.E.M.) plasma glucose (A) and insulin (B) concentrations following ingestion of four low sugar fruit drinks 2 min before consuming a mixed carbohydrate meal, in randomized order: H-BE: high blackcurrant extract; M-BE: medium blackcurrant extract; L-BE: low blackcurrant extract; CON: placebo. N=22. All data were natural log transformed before mixed model analysis. For glucose P<.001 for an overall drink effect on changes from baseline, and for insulin, P<.001 for a drink effect and P<.001 for a drink x time interaction on raw values and changes from baseline. (A) post hoc analysis of timepoint differences in change from baseline in glucose compared to CON with Dunnett's adjustment: aP<0.05 for the difference between H-BE and CON and M-BE and CON; bP<.005 for the difference between H-BE and CON; cP<0.01 for the difference between H-BE and CON. (B) post hoc analysis of timepoint differences in change from baseline in insulin with Dunnett's adjustment: aP<.005 for the difference between H-BE and CON; bP<0.01 for the difference between H-BE and CON; cP<0.05 for the difference between H-BE and CON.

Fig. 4

Postprandial plasma GIP and GLP-1 concentrations. Mean (±S.E.M.) plasma GIP (A) and GLP-1 (B) concentrations following ingestion of four low sugar fruit drinks 2 min before consuming a mixed carbohydrate meal, in randomized order: H-BE: high blackcurrant extract (GIP n=17; GLP-1 n=20); M-BE: medium blackcurrant extract (GIP n=20; GLP-1 n=22); L-BE: low blackcurrant extract (GIP n=20; GLP-1 n=21); CON: placebo (GIP n=19; GLP-1 n=22). All data were natural log transformed before mixed model analysis. For GIP, there was an overall drink effect on raw values (P<.0001) and changes from baseline (P<.0001) and a drink × time interaction on raw values (P<0.005), and for GLP-1, there was an overall drink effect on raw values and changes from baseline (P<.001). (A) post hoc analysis of timepoint differences in change from baseline in GIP with Dunnett's adjustment: aP<0.0001 for the difference between H-BE and CON and bP<0.001 for the difference between L-BE and CON; cP<0.0005 for the difference between H-BE and CON; dP<0.005 for the difference between H-BE and L-BE and H-BE and M-BE; eP<0.05 for the difference between H-BE and CON; and fP<0.0005 for the difference between H-BE and CON. (B) post hoc analysis of timepoint differences in change from baseline in GLP-1 with Dunnett's adjustment: aP<0.05 for the difference between H-BE and CON.