This article is more than 5 years old. Information may no longer be current.
Artificial pancreas system safe, effective in young children
Artificial pancreas systems appear to be safe and effective in children with type 1 diabetes aged 5 to 8 years, according to findings presented at the Endocrine Society Annual Meeting in Orlando.
“Artificial pancreas systems, or closed-loop control systems, in studies of adults and adolescents have been shown to improve mean blood sugars, but that hasn’t been done yet in young children,” Mark DeBoer, MD, MSc, MCR, associate professor at University of Virginia, said during a press conference. “There are some concerns that the systems that we have wouldn’t be able to be as useful in young children. Young children are much more sensitive to insulin, so the same amount of insulin will drop their blood sugar more, and the algorithms used in some of these artificial pancreas systems were designed for use in adults and not for young children.”
The artificial pancreas system used in this study consisted of a Tandem t:slim insulin pump, a Dexcom G4 continuous glucose monitor and a computer algorithm housed in a password-protected smart phone so the young children could not influence the input levels.
Physical activity was tracked using Fitbit devices and adjusted in the final analysis.
In this randomized, cross-over trial, the participants (n = 12; aged 5 to 8 years; median duration of diabetes, 3.7 years; mean HbA1c, 7.7%; total daily insulin, 20.3 units), received an artificial pancreas system and were followed during a monitored 68-hour camp at a ski resort. The participants were also followed for 68 hours at home using their normal diabetes care routine (insulin pump and CGM) either before or after the camp, which was decided through randomization.
The artificial pancreas system resulted in increased time with blood glucose in range (70 to 180 mg/dL; artificial pancreas = 73%; home = 46.9%, P = .002) and lower mean glucose level (artificial pancreas = 152 mg/dL; home = 190 mg/dL, P = .001) compared with the home study.
Occurrence of hypoglycemia was similar between sessions without differences in time (artificial pancreas = 6.3%; home = 20.8%).
No parent of the participants reported a child gaining access to the password that locked the smart phone screen.
“In conclusion, artificial pancreas usage was safe in this setting without dangerous lows or dangerous highs,” DeBoer said. “Clearly, we will need to test the artificial pancreas use in a home setting as we’re doing right now in adults and adolescents..
“The lock-out screens were effective in restricting access for these very young children to be able to make sure that they didn’t increase risk further” De Boer continued. “Certainly, further testing is needed to see how necessary a lock-out screen like this is and whether there are other artificial pancreas features that could be used for benefit in young children.”
Reference:
DeBoer MD, et al. OR12-2. Presented at: The Endocrine Society Annual Meeting; April 1-4, 2017; Orlando.
Disclosure: DeBoer reports no relevant financial disclosures.
Perspective
Back to Top
The use of a closed-loop system to control blood sugars has now been shown to be as efficacious in the pediatric population as it is in the adult population of people with type 1 diabetes.
Linking control of an insulin pump basal insulin infusion to information from continuous glucose monitoring systems is more efficient than using these two technologies separately in adults, and now we know the same is true for pediatrics.
Unfortunately, what is not shown is a dramatic reduction in hypoglycemia. Hypoglycemia, an event that can be both traumatic and devastating, continues to not only be the rate-limiting step in achieving normal glucose level, but also creates the dominant acute complication in caring for adolescents and children with type 1 diabetes.
This study presented an excellent “proof of concept.” The next needed step, as DeBoer stated, is to validate these studies over a longer period of time than the 68 hours of this study. The technology needs to be evaluated in real-life environments and the myriad of activities and health changes that occur in this population.
The use of this technology will undoubtedly be sought out by patients who are hopeful to remove both the extensive burden and time required to appropriately care for this disease. This study adds continued hope that the closed-loop system will offer the pediatric patients with type 1 diabetes and their families some relief from the challenges of managing this disease.
Perspective
Back to Top
To date, minimal data exists with regard to the efficacy and safety of automated insulin delivery systems in very young children. The newly available Medtronic 670G system is FDA-approved for youths aged 14 years and older, and not currently recommended for patients on less than 8 units of total daily insulin, which, until further data is published, excludes very young or small children.
In this small study of a different closed-loop system (T-slim/Dexcom/control algorithm) in youths aged 5 to 8 years, DeBoer and colleagues reported more time in a target blood glucose range of 70 to 180 mg/dL, with overall percentage of time in this range of 73%. Interestingly, the key trial of the 670G system in adolescents and adults showed a similar “time in range” of 72% (Bergenstal RM, et al. JAMA. 2016; doi:10.1001/jama.2016.11708) and in adolescents was associated with time in range (70-180 mg/dL) of 70% of all sensor values without an increase in hypoglycemia (Ly TT, et al. Pediatr Diabetes. 2016;doi:10.1111/pedi.12399).
They also report lower mean blood glucose without an increase in hypoglycemia between this closed-loop system under camp conditions in young children and usual home-based diabetes care. However, the intervention was performed in a setting (camp conditions, a ski resort, close monitoring by research staff) where we might expect lower average glycemia compared with home conditions. For parents of young children with type 1 diabetes, the fear of hypoglycemia remains understandably significant under current standards of care, while exercise remains potentially problematic with regard to automated insulin delivery, at least without an exercise signal to adjust the algorithm. Thus, to see these authors show their results regarding hypoglycemia protection under activity conditions is very encouraging.
Further studies are required to test these systems for young children, ideally in randomized controlled fashion to minimize potential confounders, such as the activity or camp setting itself. Indeed, further well-designed, larger, randomized controlled trials are required in each of the key age groups (very young children, adolescents and adults) and under specific conditions.