Feasibility of Long-Term Closed-Loop Control: A Multicenter 6-Month Trial of 24/7 Automated Insulin Delivery

Abstract

Background:

In the past few years, the artificial pancreas—the commonly accepted term for closed-loop control (CLC) of blood glucose in diabetes—has become a hot topic in research and technology development. In the summer of 2014, we initiated a 6-month trial evaluating the safety of 24/7 CLC during free-living conditions.

Research Design and Methods:

Following an initial 1-month Phase 1, 14 individuals (10 males/4 females) with type 1 diabetes at three clinical centers in the United States and one in Italy continued with a 5-month Phase 2, which included 24/7 CLC using the wireless portable Diabetes Assistant (DiAs) developed at the University of Virginia Center for Diabetes Technology. Median subject characteristics were age 45 years, duration of diabetes 27 years, total daily insulin 0.53 U/kg/day, and baseline HbA1c 7.2% (55 mmol/mol).

Results:

Compared with the baseline observation period, the frequency of hypoglycemia below 3.9 mmol/L during the last 3 months of CLC was lower: 4.1% versus 1.3%, P < 0.001. This was accompanied by a downward trend in HbA1c from 7.2% (55 mmol/mol) to 7.0% (53 mmol/mol) at 6 months. HbA1c improvement was correlated with system use (Spearman r = 0.55). The user experience was favorable with identified benefit particularly at night and overall trust in the system. There were no serious adverse events, severe hypoglycemia, or diabetic ketoacidosis.

Conclusion:

We conclude that CLC technology has matured and is safe for prolonged use in patients' natural environment. Based on these promising results, a large randomized trial is warranted to assess long-term CLC efficacy and safety.