Description:
ObjectivesThe glycemic index (GI) concept was developed by Jenkins et al in 1981 to classify the glycemic impact of carbohydrate (CHO)‐containing foods. This study aimed at quantifying inter‐ and intra‐laboratory variability on GI, Insulin Index (II), glycemic and insulinemic responses after standardising the protocols in 3 different labs.MethodsAt least 15 healthy young normal weight subjects with HOMA‐IR < 1.7 were recruited by each lab. They underwent 9 sessions to test a glucose solution (3 times) or 6 different cereal products. The 3 selected labs used the validated GI method (ISO method 26642:2010). Insulin assay and subject selection criteria were standardised.ResultsGI values for the 6 different products (mean ± SEM) ranged from 44 ± 4 to 92 ± 8. For a same product, the between‐labs differences range was 0 ‐ 11. No lab effect but significant product effects were observed for GI and iAUC glycemia. II values of the 6 different products ranged between 54 ± 3 and 85 ± 9. For a same product, the between‐labs differences range was 0 ‐ 21. Significant lab * product interaction effect was observed for II on 1 product. iAUC insulinemia displayed both significant lab and product effects. Inter‐individual and intra‐individual coefficient of variability (CV) ranges were 20 ‐ 28 % and 18 ‐ 30%, respectively for iAUC glycemia and 35 ‐ 51 % and 22 ‐ 43 %, respectively for iAUC insulinemia.ConclusionNo significant lab effect and good product discrimination was observed on GI. Insulin parameters display statistically significant differences between the labs showing the difficulty to compare the results on these parameters.