Blood glucose control in type 1 diabetes with a bihormonal bionic endocrine pancreas

Abstract

Objective: To test whether safe and effective glycemic control could be achieved in type 1 diabetes using a bihormonal bionic endocrine pancreas driven by a continuous glucose monitor in experiments lasting more than two days and including six high-carbohydrate meals and exercise as challenges to glycemic control.

Research design and methods: Six subjects with type 1 diabetes and no endogenous insulin secretion participated in two 51-h experiments. Blood glucose was managed with a bionic endocrine pancreas controlling subcutaneous delivery of insulin and glucagon with insulin pumps. A partial meal-priming bolus of insulin (0.035 units/kg/meal, then 0.05 units/kg/meal in repeat experiments) was administered at the beginning of each meal (on average 78 ± 12 g of carbohydrates per meal were consumed). Plasma glucose (PG) control was evaluated with a reference quality measurement on venous blood every 15 min.

Results: The overall mean PG was 158 mg/dL, with 68% of PG values in the range of 70-180 mg/dL. There were no significant differences in mean PG between larger and smaller meal-priming bolus experiments. Hypoglycemia (PG <70 mg/dL) was rare, with eight incidents during 576 h of closed-loop control (0.7% of total time). During 192 h of nighttime control, mean PG was 123 mg/dL, with 93% of PG values in the range of 70-180 mg/dL and only one episode of mild hypoglycemia (minimum PG 62 mg/dL).

Conclusions: A bihormonal bionic endocrine pancreas achieved excellent glycemic control with minimal hypoglycemia over the course of two days of continuous use despite high-carbohydrate meals and exercise. A trial testing a wearable version of the system under free-living conditions is justified.

Trial registration: ClinicalTrials.gov NCT01161862.

Figures

Figure 1

Mean (SD) of venous PG (A), CGMG (B), insulin and glucagon doses (C), plasma insulin levels (D), and glucagon levels (E) for all experiments. A: The mean (SD) of venous PG levels with 15-min sampling is shown from all (N = 12) 48-h experiments in six subjects. The maximum in the mean PG was 254 mg/dL at 9:15 a.m. after the first breakfast, and the mean nadir was 99 mg/dL at 12:30 p.m. at the start of the first lunch. The overall mean of all 48-h PG results (N = 193 measurements per experiment) was 158 ± 44 mg/dL. The overall mean PG during nighttime (11 p.m.–7 a.m.) was 123 ± 13 mg/dL (N = 66 measurements per experiment). The six meals are indicated by the letter M. The exercise period is indicated by Ex. B: The mean (SD) of CGMG levels, measured every 5 min, are shown. The mean peak CGMG was 229 mg/dL, recorded at 9:30 a.m. after the first breakfast, and the mean nadir was 97 mg/dL at 12:30 p.m. at the start of the first lunch. The overall mean of all CGMG measurements was 145 ± 35 mg/dL (N = 577 measurements per experiment). The overall mean CGMG overnight was 116 ± 9 mg/dL (N = 194 measurements per experiment). C: The means of all subcutaneous insulin doses (vertical lines below), including meal-priming insulin doses (indicated by downward arrows), and glucagon doses (vertical lines above), administered by the bionic endocrine pancreas, are indicated. The mean daily doses of insulin and glucagon administered by the program were 0.5 units/kg/day and 3.64 μg/kg/day, respectively. D: The mean (SD) plasma insulin levels, measured every 30–60 min, with mean over 48 h of 38 ± 10 μIU/mL. The tmax for insulin absorption ranged from 24–166 min in all subjects and was 70 min on average. E: The mean (SD) plasma glucagon levels, measured every 30–60 min. The mean glucagon level over 48 h was 83 ± 28 pg/mL, with peak mean levels increasing only transiently over the normal fasting range (indicated by shaded area) to 180 pg/mL at 5:00 p.m. after exercise. The peak level is consistent with the increased glucagon dosing at the time of exercise seen in (C).

Figure 2

Cumulative distributions for 48-h and night PG profiles. The solid curve shows the distribution of all recorded venous PG levels (N = 2078) in the twelve 48-h closed-loop experiments. Venous PG was <70 mg/dL 0.7% of total study time, within 70–120 mg/dL 38% of time, and within the American Diabetes Association target range of 70–180 mg/dL 68% of the time. The dashed curve shows the distribution of all night (11:00 p.m.–7:00 a.m.) venous PG levels (676 measurements over 2 nights for each experiment). Venous PG was <70 mg/dL 0.5% of total study time, within 70– 120 mg/dL 62% of the time, and within 70–180 mg/dL 93% of the time. A previous meta-analysis of closed-loop control studies limited to the night hours reported 76% of PG values in adults with type 1 diabetes were within their specified target range of 70–144 mg/dL (11). In this study, 84% of venous PG values were within this range despite one of the insulin delivery failures occurring during night hours (Supplementary Fig. 7).

Figure 3

Glucose profiles and the corresponding insulin–glucagon doses that were administered by the control algorithm are shown for all eight hypoglycemic episodes. A–C: Hypoglycemic episodes (defined as events with PG <70 mg/dL) that occurred during experiments using 0.05 units/kg meal-priming boluses (venous PG shown as stars and CGMG as circles). D–F: Hypoglycemic episodes that occurred during experiments that used 0.035 units/kg meal-priming boluses. The isolated PG value of 69 mg/dL in (A) appears to be a measurement artifact. The first PG value <70 mg/dL in each of the five hypoglycemic episodes shown in (B–E) were 28, 35, 47, 48, and 89 mg/dL lower than their corresponding CGMG values, which were 94, 88, 107, 97, and 148 mg/dL, respectively. Both interventions in (F) were at the request of the subject and, in both cases, PG was >60 mg/dL and CGMG was >80 mg/dL. Note that in (D) the meal was presented to the subject 20 min late, at 18:20, although the meal-priming bolus was administered on time, at 18:00. (tmax for that experiment was 49 min.)