British Journal of Nutrition

Research Article

A randomised four-intervention crossover study investigating the effect of carbohydrates on daytime profiles of insulin, glucose, non-esterified fatty acids and triacylglycerols in middle-aged men

Audrey E. Brynesa1, C. Mark Edwardsa2, Mohammed A. Ghateia2, Anne Dornhorsta2, Linda M. Morgana3, Stephen R. Blooma2 and Gary S. Frosta1 c1

a1 Nutrition and Dietetic Research Group, Hammersmith Hospital, Du Cane Road, London W12 0HS, UK

a2 Endocrine Unit, Imperial College School of Medicine, Hammersmith Hospital, Du Cane Road, London W12 OHS, UK

a3 School of Biomedical and Life Sciences, University of Surrey, Guildford GU2 7XH, UK


Postprandial concentrations of glucose, insulin and triacylglycerols (TG) correlate to risk for CHD. Carbohydrates affect many metabolites that could have a potential effect on cardiovascular risk factors. The objective of the present study was to examine, using a randomised prospective study, the acute (day 1) and ad libitum medium-term (day 24) effects of four diets: a high-fat diet (HIGH-FAT; 50 % fat, >34 % monounsaturated fatty acids); a low-glycaemic index (GI) diet (LOW-GI; high-carbohydrate, low-GI); a high-sucrose diet (SUCROSE; high carbohydrate increase of 90 g sucrose/d); a high-GI diet (HIGH-GI; high-carbohydrate, high-GI). Daytime profiles (8 h) (breakfast, lunch and tea) of lipid and carbohydrate metabolism were completed during day 1 and day 24. Seventeen middle-aged men with one or more cardiac risk factors completed the study. There was no change from day 1 or between diets in fasting glucose, lipids or homeostatic assessment model (HOMA) on day 24. The HIGH-FAT compared with the three high-carbohydrate diets was associated with lower postprandial insulin and glucose but higher postprandial TG and non-esterified fatty acids (NEFA). There was a significant increase in the 6 h (15.00 hours) TG concentration (day 1, 2·6 (SEM 0·3) MMOL/L v. DAY 24, 3·3 (sem 0·3) mmol/l; P<0·01) on the SUCROSE diet. Postprandial HOMA (i.e. incremental area under the curve (IAUC) glucose (mmol/l per min)×IAUC insulin/22·5 (mU/l per min)) median changes from day 1 to day 24 were −61, −43, −20 and +31 % for the HIGH-FAT, LOW-GI, SUCROSE and HIGH-GI diets respectively. The HIGH-GI percentage change was significantly different from the other three diets (P<0·001). Despite being advised to maintain an identical energy intake there was a significant weight change (−0·27 (sem 0·3) kg; P<0·02) on the LOW-GI diet compared with the SUCROSE diet (+0·84 (sem 0·3) kg). In conclusion the HIGH-FAT diet had a beneficial effect on postprandial glucose and insulin over time but it was associated with higher postprandial concentrations of TG and NEFA. Conversely the HIGH-GI diet appeared to increase postprandial insulin resistance over the study period.

(Received March 22 2002)

(Revised August 17 2002)

(Accepted August 22 2002)


c1 *Corresponding author: Dr Gary Frost, fax +44 20 8383 3379, email