Feasibility of a Bihormonal Closed-Loop System to Control Postexercise and Postprandial Glucose Excursions

Background: The aim of this pilot study was to test the feasibility of a bihormonal (glucagon and insulin) closed-loop (CL) system by challenging the system with two meals and 30 min exercise. Methods: Ten patients with type 1 diabetes treated with continuous subcutaneous insulin infusion underwent a standardized protocol on three different occasions: 40 g carbohydrate breakfast followed 2 h later by 30 min of moderate-intensity exercise, followed 1.5 h later by a standardized 60 g carbohydrate lunch. An open-loop (OL) day served as control, the first CL day as tuning experiment, and the second CL day to compare with OL. Results: The overall mean venous glucose was similar: 9 (5.4–13.5) mmol/liter in OL versus 8.7 (6.4–11.0) mmol/liter in CL, p = .74. The postbreakfast glucose concentrations tended to be lower in OL than in CL [9.5 (4.3–13.3) versus 11.4 (7–16.2) mmol/liter; p = .07] and higher in OL than in CL postlunch [9.4 (6.0–14.9) versus 7.7 (5.5–9.0) mmol/liter, p = .15]. The postexercise glucose concentrations were similar in OL and CL: 7.5 (4.6–13) versus 8.2 (5.5–13.1) mmol/liter; p = .45. In those patients coming in with baseline glucose above 7 mmol/liter, there was initial overinsulinization in CL. During OL, two hypoglycemic episodes occurred compared with four hypoglycemic episodes in three participants during CL. Glucagon seemed mostly effective in preventing hypoglycemia. Conclusions: Overall, CL glucose control was comparable to OL control, but there was overinsulinization in those patients with baseline glucose above 7 mmol/liter.