Sensor-Augmented Pump Therapy in Very Young Children with Type 1 Diabetes: An Efficacy and Feasibility Observational Study

Introduction

I n children with type 1 diabetes (T1D), optimal glycemic control enhances growth and ensures normal pubertal development and emotional maturation. Moreover, tight glycemic control delays long-term complications of T1D, often at the cost of increasing the risk for hypoglycemia.

Continuous glucose monitoring (CGM) provides frequent glucose readings, and their accuracy and reliability have been established in both adults ^1,2 and children. ³

Recent studies demonstrated the advantages of sensor-augmented pump (SAP) therapy both in randomized controlled trials ⁴ and in real-life setting, ⁵ with a mean decrease of 0.5–0.6% in glycosylated hemoglobin (HbA1c) levels, without increasing the risk of hypoglycemia. Nevertheless, there are limited data on its use in very young children. ⁶ Moreover, guidelines and recommendations have been recently published for both adults ⁷ and children. ⁸

The aim of the present study was to evaluate the efficacy and feasibility of daily CGM in very young children with T1D in a real-life setting.

Subjects and Methods

This is a multicenter retrospective study that involved three pediatric diabetologic centers belonging to the Study Group on Diabetes of the Italian Society of Pediatric Endocrinology and Diabetology. They are among the referral centers for the diagnosis and treatment of pediatric diabetes and are located in Northern Italy.

Data were analyzed only regarding patients with T1D 7 years of age or younger, using insulin pump (Animas [West Chester, PA] model IR1200 or 2020) therapy for 6 months or more and using CGM (Dexcom [San Diego, CA] Seven Plus™).

For each case the following data were collected from medical records: sex, the age at the last visit, the age at which the pump therapy began, age at which the CGM began, metabolic control evaluated as value of HbA1c (measured at each center using a model DCA-2000 analyzer [Siemens/Bayer, Milan, Italy]), reason to start CGM, and hypoglycemic and diabetic ketoacidosis episodes. Severe hypoglycemia was defined as a blood glucose measurement of <70 mg/dL (3.9 mmol/L) with a loss of consciousness or the patient's need for assistance. Data collection was completed by August 2011.

For each patient identified using both insulin pump and CGM (SAP therapy) and his or her parents we rated satisfaction, ease of use, ease of learning, comfort, and safety using a suitable rating scale with 1 meaning low grading, 2 average grading, and 3 high grading.

The primary end point was the change in HbA1c between baseline (start of using CGM) and the end of the observation period. Episodes of severe hypoglycemia and diabetic ketoacidosis, days of sensor use, and its feasibility were evaluated too.

The study was conducted according to the Declaration of Helsinki, and both parents and physician signed an informed consent about using CGM at such a young age. ⁹ Indeed, CE regulation allows the use of the Dexcom Seven Plus system only at 18 years of age or older.

Continuous variables are displayed as frequencies or percentages. The paired t test was used to analyze changes of continuous variables over time. Statistical significance was determined for a P value of 0.05.

Results

SAP therapy has been identified in 28 children (15 boys, 13 girls), 3–7 years old (mean±SD, 6.5±1.2 years), with T1D since 3.6±1.5 years (Table 1).

Before starting SAP therapy children had been using an insulin pump since 2.2±1.2 years; SAP therapy has been continued for 0.8±0.5 years, with nine patients discontinuing using the sensor because of difficult child compliance (n=4), skin reactions (n=2), and sensor malfunction (n=3), throughout the follow-up (five patients discontinued using the sensor by 2.2±0.5 months, while the other four patients discontinued using the sensor at the end of the observational period, mostly because of problems with sensor funding by the health system). Patients wore sensors for 83% of the time, with 18 of them (64%) using the sensor all the time. Mean sensor life has been 12±0.5 days (range, 7–21 days).

A slight but not significant difference of −0.2% in HbA1c has been observed at the end of the study versus baseline (7.3±0.8% vs. 7.1±0.5%, P not significant). However, stratifying the patients according to baseline HbA1c (>7.5% vs. ≤7.5%), a significant improvement (−0.9%) has been observed in patients with HbA1c >7.5% (8.4±0.8% vs. 7.5±0.4%, P=0.026). Excluding from the analysis patients who discontinued using the sensor both during the follow-up or at the end, no difference in P values has been observed (P not significant considering all patients, P=0.021 considering patients with baseline HbA1c >7.5%). For this reason we considered for the statistical analysis all patients who discontinued and who did not (indeed, some of the patients were “forced” to discontinue because of problems with sensor funding, and others switched to a different system [i.e., Paradigm Veo™ from Medtronic, Northridge, CA]).

No difference has been observed regarding diabetic ketoacidosis and severe hypoglycemia before and after the follow-up, even if these data would not be meaningful in such a small sample size over this short time period. Moderate hypoglycemia has been reduced, even if not significantly, between baseline and the end of follow-up (12 episodes/100 patients/year vs. 8.5 episodes/100 patients/year, P not significant).

Judgment regarding device comfort has been positive (score of 3, 62%), as well as ease of learning (score of 3, 73%), and usefulness for daily diabetes management (score of 3, 69%). The greatest perceived benefit was the reduced fear of hypoglycemia and usefulness of alarms (score of 3, 81%). Combining the insulin pump with the sensor has been really appreciated by both the patients and the parents (overall satisfaction, 92%; score of 3, 85%; score of 2, 7%).

Discussion

SAP (Dexcom Seven Plus system) therapy plus an insulin pump seems to be effective and feasible even in preschool children with T1D, in a real-life setting. To our knowledge, this is the first study ever in which SAP has been investigated in a quite large group of children under 7 years of age in a real-life setting. After 10 months using SAP, HbA1c improved only slightly, but in the subgroup with baseline HbA1c >7.5%, a significant improvement has been observed with a 0.9% reduction in HbA1c, in line with that observed in the 6–18-year-old group in the STAR 3 study at the 9-month time point of the follow-up. ⁴ However, it is noteworthy to see that mean HbA1c at baseline was good and has been kept steady during the observation period. The only study published on this topic just recently ⁶ in a much larger group of children under the age of 10 years failed to observe any improvement in HbA1c in the whole group, even in the subgroup with HbA1c >8.0% at baseline (−0.3% decrease). The reason for this difference is not clear, but probably due to the lesser fear of hypoglycemia observed in our group, as stated by the rating scale results (81% of parents recorded the highest score about the fear of hypoglycemia), which is in contrast with what has been observed in the study by Mauras et al. ⁶

The SAP involving the Dexcom Seven Plus has been accepted and appreciated by both the children and the parents with a great overall satisfaction. We did not use any quality of life questionnaire, but our results seem similar to what was observed in the STAR 3 trial and recently published. ¹⁰ It is interesting to note that managing two tools (insulin pump and CGM system) did not affect the satisfaction of both the children and the parents, even in patients so young.

Further studies involving a greater study population are necessary in order to eventually establish age-specific behavioral algorithms, which may help reducing the glycemic excursions so often seen in the preschool age.

In conclusion, SAP therapy in very young children seems to be safe and effective, with the greatest benefit observed in patients with higher HbA1c at baseline (>7.5%).