Estimation of Meal-Time Insulin Based on Glycemic Load versus Carbohydrate Counting among Patients with Type 1 Diabetes Using Insulin Pump

Abstract Background Dietary carbohydrates have a major effect on postprandial glycemic response. Therefore, the current guidelines recommend the usage of insulin/carbohydrate ratio for calculation of prandial insulin dose based on the carbohydrate count in a given meal. However, some studies pointed to the importance of the quality of carbohydrates, as reflected by their glycemic index, in addition to their quantity on the regulation of postprandial glucose level. Objectives To identify a safe, feasible and effective method for calculation of prandial insulin dose based on the glycemic load of a meal. This could minimize postprandial glycemic variability in type 1 diabetic patients. Patients and Methods Our study included 15 type 1 diabetic participants; their ages ranged from 8 to 15 years. The participants were given a breakfast with the same amount of fat, protein and carbohydrates, which was prepared similarly in all studied patients. Sequence, quality and amounts of meals consumed during the first week of the study were replicated during the second week. Calculation of prandial insulin doses using one regimen for 1 week and a crossover to the other regimen for an additional week was done. A detailed medical history was collected, including the age of onset of diabetes, disease duration, and symptoms of diabetic complications. Results Blood glucose postprandial excursions was significantly lower during the glycemic load counting (GLC) than the carbohydrate counting (CC) period from baseline to 120 min (-10.20 ± 8.18 vs 8.47 ± 6.65 mg/dl, respectively, p = < 0.001), baseline to 180 min (-11.80 ± 8.16 vs 5.13 ± 7.01 mg/dl, respectively, p = < 0.001) and baseline to 240 min (-9.60 ± 6.07 vs -0.53 ± 7.31 mg/dl, respectively, p = 0.007). However, there was no postprandial glucose difference between the two periods from baseline to 60 min (-2.87 ± 8.09 vs 2.53 ± 7.76 mg/dl, respectively, p = 0.083). Conclusion The data from this study showed that glycemic load counting for prandial insulin doses is a feasible and effective method in short term glucose control.