Dr. Jennie C. Brand-Miller, from the University of Sydney stated that, “The food insulin index (FII) may provide a better way to adjust insulin dose in Type 1 diabetes…. In time, it may also enable us to design diets to prevent diabetes.”
Dr. Brand-Miller and colleagues sought to expand their FII database (previously containing 38 foods) and to determine whether the concept was able to predict insulin responses to mixed meals composed of variable amounts of foods whose individual FII had been previously determined.
In the study, 21 healthy subjects consumed 13 isoenergetic mixed meals in 2000 kJ portions with varying nutrient content. The insulin responses were strongly correlated with calculated insulin demand predicted by the FII of the component foods (r=0.78), the authors report.
The calculated glycemic load was less strongly correlated with the observed insulin response (r=0.68), and the carbohydrate content was not a significant predictor of average insulin response.
Fat content was inversely associated with the observed insulin response, whereas protein and fiber content showed no relationship with insulin demand, according to the researchers.
“We have completed an 8 hour study in healthy subjects showing that the FII works well in predicting both higher and lower day-long insulin levels at breakfast, morning tea, and lunch,” Dr. Brand-Miller said. “We are about to start studies in individuals with Type 1 diabetes who use an insulin pump. We predict that adjustments to insulin dose with the FII will provide better blood glucose control than carbohydrate counting.”
“We are also looking at large cohort studies … to see if people who choose diets with a higher insulin demand have higher rates of Type 2 diabetes than those who choose a diet with a lower insulin demand,” Dr. Brand-Miller added.
She pointed out, however, that the FII “may not predict actual insulin levels in Type 2 diabetes because (the) pancreas is failing.”
Am J Clin Nutr 2009;90:986-992.