Use of an artificial pancreas as a tool to determine subcutaneous insulin doses in juvenile diabetes

The present study was undertaken to examine the feasibility of determining the most appropriate subcutaneous insulin treatment in unstable diabetes on the basis of the circadian hormonal profile delivered by an artificial pancreas. The metabolic control of 11 brittle diabetic subjects, as assessed by the M value and the MAGE index (used as indexes of blood glucose control and of glycemic fluctuations, respectively), was compared during a 5-day period before and after a 24-h connection to the artificial pancreas. The usual insulin treatment was continued to that day. Examination of the insulin pattern revealed by the artificial pancreas suggested that a valid scheme for subsequent treatment should consist of two daily injections of a mixture of short-acting and intermediate-acting insulins, which was administered to the patients beginning with the injection given after the artificial pancreas onwards. The new insulin regimen was characterized by a total daily dose that increased from 0.93 +/- 0.10 to 1.20 +/- 0.10 U/kg body weight (mean +/- SEM; P less than 0.005) as well as by a higher proportion of the dose given as regular insulin (37.1 +/- 6.9% before vs. 56.0 +/- 2.1% after; P less than 0.05). These changes led to a better control of blood glucose in 10 patients, as evidenced by a decrease of both the M value and the mean of all blood glucose levels. The mean MAGE index was not decreased, however, by the new insulin program, thereby suggesting that the lability of the disease remained unabated. These results indicate that subcutaneous treatment consisting of two daily injections of regular and intermediate-acting insulins and comprising 50 to 60% of the former could improve the metabolic control in unstable diabetes. The artifical pancreas provided a rapid and simple means to determine the appropriate doses for each type of insulin.