Estimating insulin demand for protein-containing foods using the food insulin index

Abstract

Background/objective: The Food Insulin Index (FII) is a novel algorithm for ranking foods on the basis of insulin responses in healthy subjects relative to an isoenergetic reference food. Our aim was to compare postprandial glycemic responses in adults with type 1 diabetes who used both carbohydrate counting and the FII algorithm to estimate the insulin dosage for a variety of protein-containing foods.

Subjects/methods: A total of 11 adults on insulin pump therapy consumed six individual foods (steak, battered fish, poached eggs, low-fat yoghurt, baked beans and peanuts) on two occasions in random order, with the insulin dose determined once by the FII algorithm and once with carbohydrate counting. Postprandial glycemia was measured in capillary blood glucose samples at 15-30 min intervals over 3 h. Researchers and participants were blinded to treatment.

Results: Compared with carbohydrate counting, the FII algorithm significantly reduced the mean blood glucose level (5.7±0.2 vs 6.5±0.2 mmol/l, P=0.003) and the mean change in blood glucose level (-0.7±0.2 vs 0.1±0.2 mmol/l, P=0.001). Peak blood glucose was reached earlier using the FII algorithm than using carbohydrate counting (34±5 vs 56±7 min, P=0.007). The risk of hypoglycemia was similar in both treatments (48% vs 33% for FII vs carbohydrate counting, respectively, P=0.155).

Conclusions: In adults with type 1 diabetes, compared with carbohydrate counting, the novel FII algorithm improved postprandial hyperglycemia after consumption of protein-containing foods.