(P06-002-24) Equivalent Glycemic Load and Insulinemic Response Elicited by 7 Low-Carbohydrate Products in Healthy Adults

Objectives: Equivalent glycemic load (EGL) is a method to quantify glycemic impact of low-carbohydrate foods and is defined as the grams (g) of available-carbohydrate (avCHO) from white-bread that elicits the same glycemic response as 1 serving of a low-carbohydrate food. The objectives of this study were to test the hypothesis that the EGL measurement is accurate and precise to within ±1g, to measure the EGL of 7 low-carbohydrate food products and to determine if their nutritional composition is related to their postprandial glucose and insulin responses.

Methods: A randomized, crossover, open-label study was performed with n=25 healthy adults. On 10 separate mornings, overnight fasted subjects consumed either a portion of white-bread containing either 20.8g avCHO (WB20) or 5.2g avCHO (WB5) or one of the 7 test food products; each subject tested WB20 twice. Plasma glucose was measured before and at intervals for 2 hours after starting to eat; serum insulin was measured in a subset of n=10 subjects. EGL is defined as T/(WB/20.8) where T is the plasma glucose incremental area-under-the-curve (iAUC) after the test food, and WB is the mean iAUC after WB20.

Results: The expected EGL value of WB5 was 5.2g, the mean measured value was 4.0g (95% margin of error = 0.6g). The EGLs of all 7 test food products were significantly less than WB5 (p < 0.05), ranging from 0.6 to 2.4g. Among test food products, the only significant correlation between nutritional content and postprandial response was between g protein/g avCHO and insulin iAUC (r=0.86, p=0.014).

Conclusions: The results support the hypothesis that the EGL measure is accurate and precise to within approximately ±1g. All 7 low-carbohydrate food products elicited a small glycemic response, less than a portion of white bread containing 5.2g avCHO. Finally, the results suggest that the glycemic impact of low-carbohydrate foods cannot be predicted from their nutritional composition, but their insulinemic responses are positively related to protein per g avCHO.

Funding Sources: Funding for this study was provided by the Bell Institute of Health and Nutrition, General Mills. TB and KM were employed by General Mills at the time of the study. TW was an employee of INQUIS Clinical Research. Neither he nor his wife have any financial interest in General Mills.