Home / Resources / Articles / Dual-Hormone Closed-Loop Delivery System Compatible with Exercise

Dual-Hormone Closed-Loop Delivery System Compatible with Exercise

Jul 28, 2018

Dual-Hormone Closed-Loop Delivery System: addition of glucagon prevents exercise-induced hypoglycemia.

Automated insulin delivery systems are becoming the new standard of treatment for type 1 diabetes. With an insulin pump and glucose sensor, these automated systems can correct dysglycemia and mimic the body’s pancreas. However, exercise remains an issue in many of these patients as it often leads to hypoglycemia. As we work out, our muscles consume glucose for energy and insulin sensitivity increases, leading to an increased risk of hypoglycemia. The thought of a dual hormone closed-loop system consisting of insulin and glucagon may be a solution for this issue. In an inpatient study, the use of a dual hormone system was compared to a single hormone system that both adapted to exercise when announced at the start of exercise.

At Oregon Health & Science University, 25 people with type 1 were enrolled and randomized to four arms of the study using Latin squares design in blocks of four. Inclusion criteria for the study included diagnosis of type 1 diabetes for at least one year, age range of 21 years old, use of an insulin pump, and the ability to perform at least 45 minutes of exercise minimum. Participants were also required to live with another adult. Exclusion criteria included a past medical history of cardiovascular disease, kidney disease, liver disease, or unmanaged hypertension. Pregnancy, severe hypoglycemia, hypoglycemia unawareness, corticosteroid use, immunosuppressant use, and seizure disorders were also included in the exclusion criteria. Of the 25 patients who were screened, 20 were eligible to enter the study.

The participants were trained for two weeks and tested the Tandem slim pumps and Dexcom continuous glucose monitor (CGM) before beginning the four study arms. In a randomized order, participants tested the dual-hormone closed-loop system, the single-hormone closed-loop system, a predictive low glucose suspend system, and continuation of current treatment. The participants wore the Dexcom CGM and sensors were placed 24 to 72 hours prior to intervention and were then calibrated every 12 hours. On day 1 and day 4 of each study arm, participants stayed at the clinic for 12 hours and chose meals for breakfast, lunch, and dinner. The same meals were given on day 4. Participants resumed regular daily activities and were then expected to perform aerobic exercise on a treadmill at 2 pm each day. CGM readings were transmitted to the study phone via Bluetooth and were blinded to the patients during this care. If glucose levels fell below 70 mg/dL, participants were given 15 grams of carbohydrates to consume.

The insulin provided in each pump was insulin aspart and the glucagon provided in the dual systems was Glucagen that was reconstituted every 24 hours. Alarms were sent off if the glucose levels fell below 40 mg/dL or if they elevated to 400 mg/dL or more. Exercise detection occurred when the metabolic equivalent to task (METs) exceeded 4 for 5 or more sequential minutes. insulin was then turned off for 30 minutes and then reduced by 50% for 60 minutes. in addition to insulin adjustments, the target glucose level provided by glucagon was 95 to 120 mg/dL in the dual hormone system. The dual system was adaptive so that if the participant experienced hypoglycemia in the first 1.5 hours, glucagon would be released sooner after exercise is detected to provide a glucose level of 120 to 130 mg/dL. For the predictive suspension algorithm, a Dexcom CGM and Tandem insulin pump was also used and designed to suspend insulin delivery when glucose levels ranged from 70 to 140 mg/dL and predicted to drop below 90 mg/dL in 30 minutes. Primary outcomes were the percentage of time the participant was hypoglycemic (<70 mg/dL) and the time in range (70 to 180 mg/dL).

The dual-hormone system had the lowest percentage of time in hypoglycemia, 3.4% compared to the single hormone system 8.3% (p <0.009). The predictive suspend system percentage was 7.6% (p<0.001). The only adverse event was gastrointestinal upset. No events were classified as serious. Because the participants in the current treatment group could adjust their insulin before they worked out, they had lower rates of hypoglycemia. Automated systems were developed for those patients who often forget to make these adjustments. To perfect this new method of type 1 diabetes care, glucagon needs to be released in doses that prevent hypoglycemia but that will not cause hyperglycemia and side effects such as nausea. In conclusion, the dual hormone system enables patients to exercise with less hypoglycemia compared to the standard single-hormone closed-loop system.

Practice Pearls:

  • Automated systems are becoming the new standard treatment for people with type 1 diabetes.
  • Single-hormone closed-loop system and self-administration of insulin can result in exercised-induced hypoglycemia.
  • Dual hormone closed-loop system has shown to be superior at preventing hypoglycemia than single-hormone closed-loop systems during exercise.
  • Glucagon doses need to be adjusted appropriately to minimize hyperglycemia and adverse events.


Castle, Jessica R., et al. “Randomized Outpatient Trial of Single- and Dual-Hormone Closed-Loop Systems That Adapt to Exercise Using Wearable Sensors.” Diabetes Care, vol. 41, no. 7, Nov. 2018, pp. 1471–1477., doi:10.2337/dc18-0228.