High-Fiber Cereal Reduces Postprandial Insulin Responses in Hyperinsulinemic but not Normoinsulinemic Subjects

Wolever, Thomas M.S. DM, PHD
Campbell, Janice E. MSC
Geleva, Daniela PHD
Anderson, G. Harvey PHD

From the ¹Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada; the ²Division of Endocrinology and Metabolism, St. Michael's Hospital, Toronto, Ontario, Canada; and ³General Mills, Minneapolis, Minnesota.

Address correspondence and reprint requests to Thomas MS Wolever, DM, PhD, Department of Nutritional Sciences, 150 College St., Room 316, Toronto, Ontario, Canada M5S 3E2. E-mail: [email protected].

Received for publication 5 December 2003 and accepted in revised form 20 February 2004.

T.M.S.W. is employed by Glycaemic Index Testing and has received grant support from General Mills. J.E.C. is employed by Glycaemic Index Testing. G.H.A. has received grant support from and is on an advisory committee for General Mills and has received honoraria from General Mills, Heinz, McDonalds, and the Canadian Sugar Institute.

Abbreviations: AUC, area under the curve; FPI, fasting plasma insulin; GIP, gastric inhibitory polypeptide; GLP-1, glucagon-like peptide-1; IGT, impaired glucose tolerance; RGR, relative glucose response; RIR, relative insulin response.

A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.

Diabetes Care 27(6):p 1281-1285, June 2004.

OBJECTIVE

The objective of this study was to compare the plasma glucose and insulin responses elicited by two ready-to-eat breakfast cereals (one being high and the other being low in cereal fiber) and to see if the differences in response depended on subjects' fasting plasma insulin.

RESEARCH DESIGN AND METHODS

Nondiabetic men (n = 77) were studied on two occasions after 10- to 14-h overnight fasts. They consumed 25 g of available carbohydrate from high- or low-fiber breakfast cereals in random order with blood taken at intervals for 2 h. Data from the 42 men with high fasting plasma insulin (FPI) on screening (>40 pmol/l) were compared with those from the 35 men with normal FPI (≤40 pmol/l).

RESULTS

Hyperinsulinemic men had significantly higher waist circumference and BMI, lower HDL cholesterol, and a trend toward higher triglycerides (P = 0.07) than control subjects. In all 77 subjects, the incremental area under the glucose response curve (AUC) after high-fiber cereal was 11.8 ± 5.5% (P = 0.036) less than after low-fiber cereal with the reductions being equivalent in the hyperinsulinemic (12.6 ± 8.3%) and control (10.9 ± 9.1%) groups. However, insulin peak rise was reduced by the high-fiber cereal only in hyperinsulinemic men (351 ± 29 vs. 485 ± 55 pmol/l) but not in control subjects (211 ± 20 vs. 220 ± 20 pmol/l; cereal × group interaction P = 0.044). Insulin AUC after the high-fiber cereal, expressed as a percentage of that after low-fiber cereal, was negatively related to FPI (P = 0.009) but not to age, BMI, or waist circumference.

CONCLUSIONS

The high-fiber cereal reduced glucose responses to the same extent in normal and hyperinsulinemic men, but reduced insulin responses only in hyperinsulinemic subjects.