Page 325 - Glucose Monitoring Devices

332 CHAPTER 16 The dawn of automated insulin delivery

delivery when the sensor glucose is predicted to be below 20 mg/dL (1.1 mmol/L)
above the preset threshold within 30 min. Thus, a predicted sensor glucose of
85 mg/dL (4.9 mmol/L) in the next 30 min would trigger an insulin suspension.
Sixty-nine participants completed the study, and hypoglycemia was prevented in
60% of the instances [30]. Mean sensor glucose at the time of suspension was
101 mg/dL (5.6 mmol/L), and the mean duration of the suspension was 105 min.
Two hours after the start of the suspension, sensor glucose levels reached 102 mg/
dL (5.7 mmol/L) [30]. In a 6-month multicenter randomized controlled trial,
children and adolescents with T1D were randomized to the 640G system with
PLGS or SAP therapy. The results showed a significant reduction in hypoglycemia
(sensor glucose < 63 mg/dL) with PLGS (1.5% vs. 2.6%) without glycemic deteri-
oration as measured by HbA1c levels [31].
In a study by Wood and colleagues, the MiniMed 670G “suspend before low”
feature was tested in children with T1D aged 7e13 in an overnight in-clinic study,
using a preset limit of 65 mg/dL (3.6 mmol/L) [32]. This PLGS system was able to
prevent hypoglycemia 80% of the time after exercise, without causing rebound
hyperglycemia [32]. Additionally, there were no severe hypoglycemic events
reported [32].
The t:slim X2 pump with Basal IQ allows for basal insulin suspension if a sensor
glucose of less than 80 mg/dL (4.4 mmol/L) is predicted in the next 30 min based on
CGM trends and calculated insulin on board or if the sensor glucose falls below a
threshold of 70 mg/dL (3.9 mmol/L). Basal insulin resumes when the sensor read-
ings are higher than the previous reading if the sensor glucose is no longer predicted
to go below 80 mg/dL (4.4 mmol/L), if sensor data are lost for 10 min, or if insulin
suspension is longer than 120 min in any 150-min period of time. Approval of this
device was based on data collected in the PROLOG trial, which was a 6-week,
outpatient, randomized crossover study including 103 participants and comparing
the t:slim X2 pump with Basal IQ PLGS system to SAP therapy [33]. The use of
the system was found to lead to a 31% reduction in time less than 70 mg/dL
(3.9 mmol/L) without an increase in time spent with sensor glucose readings over
180 mg/dL (10 mmol/L) [33]. Participants in the trial found the system to be “excep-
tional” regardless of their previous experience with diabetes technology when
queried with a usability scale [34].
Just as concerns existed regarding the risk of ketosis with an LGS system, the
same held true for PLGS systems. Therefore, data were analyzed from an outpatient,
home-based study of 45 adolescents and adults using a PLGS system to assess the
frequency of morning ketosis after overnight insulin suspension [35]. The system
suspended basal insulin delivery on 76% of the 977 study nights [35]. On the
mornings following an overnight basal insulin suspension, blood ketones were found
to be greater than 0.6 mmol/L 1.5% of the time [35]. When limiting the analysis to
nights when insulin suspension exceeded 2 h, only 2 of the 159 nights (1.3%) had
subsequent morning ketone levels greater than or equal to 0.6 mmol/L [35]. This
study demonstrated the safety of predictive basal insulin suspension and concluded
that routine measurements of blood or urine ketones were not necessary [35].