PracticeUpdate: Endocrinology | Volume 1. No 3. 2016

15 / 16

spent in the hypoglycaemic range between
the two study conditions (P = 0.28).
Continued use of the closed-loop system
was offered to participants at the end of
the randomised trial, with 14 participants
(~90%) choosing this option. The median
time in the continuation phase was ~11.6
weeks, and mean glucose achieved during
this period was 126 mg/dL (6.99 mmol/L),
with 68.7% of time in target. This continu-
ation phase encompassed use of the system
both during labour and after delivery. Time
in target was 86.8% in the 24 hours prior
to delivery and 73.7% in the 48 hours post-
delivery. The system adeptly adjusted insulin
delivery, reducing it 53% from pre-delivery
doses in the post-delivery period.
As maternal hyperglycaemia impacts fetal
growth, the achievement of more targeted
control through closed-loop use has the
potential to minimise excessive weight
gain. The vast majority of infants in the
present study were categorised as large for
gestational age. With no change in rates of
hypoglycaemia and no episodes of severe
hypoglycaemia in either study condition, it
may be feasible to make the algorithm more
aggressive in future studies.
Stewart and colleagues have demon-
strated that, in pregnancy, when insulin
requirements frequently change and there is
substantial day-to-day glycaemic variability,
the closed-loop system can help expectant
mothers achieve more targeted glycaemic
control. Importantly, this was achieved
without any changes in the programming
to the system. With the clear-cut benefits
of closed-loop therapy in this population,
the next steps should be application of such
technologies prior to conception with use
extending for the duration of the pregnancy
to further determine how it can impact both
maternal and neonatal outcomes.
References
1. Ruan Y, Elleri D, Allen JM, et al. Diabetologia
2015;58:687-690.
2. Dauber A, Corcia L, Safer J, et al. Diabetes Care
2013;36:222-227.
3. Murphy HR, Elleri D, Allen JM, et al. Diabetes Care
2011;34:406-411.
4. Murphy HR, Kumareswaran K, Elleri D, et al. Dia-
betes Care 2011;34:2527-2529.
5. Stewart ZA, Wilinska ME, Hartnell S, et al. N Engl
J Med 2016;375:644-654.
Dr Sherr is Instructor, Pediatrics
(Endocrinology), Yale University School
of Medicine, New Haven, Connecticut.
Her research has focused on artificial
pancreas technologies and the use
of adjunctive therapy to mitigate
postprandial hyperglycaemia.
In pregnancy, when insulin requirements frequently change and there is
substantial day-to-day glycaemic variability, the closed-loop system can help
expectant mothers achieve more targeted glycaemic control. Importantly, this
was achieved without any changes in the programming to the system.
Extending on their previous work, the
study “Closed-loop insulin delivery
during pregnancy in women with type
1 diabetes” by Stewart and colleagues was
an open-label, randomised crossover assess-
ment examining overnight glucose control
following 4 weeks of closed-loop therapy
compared with 4 weeks of sensor-augmented
pump therapy (SAP).5 The system used was
the Florence D2W closed-loop system. The
system is comprised of a FreeStyle Naviga-
tor II that was linked via a cable to a Dell
Latitude 10 tablet, which housed the model
predictive control algorithm that commu-
nicated via Bluetooth with the DANA
Diabecare R insulin pump to deliver insulin
based on algorithm calculations every 12
minutes. The treat-to-target algorithm was
set with glucose targets between 97 and
124 mg/dL (5.38 and 6.88 mmol/L). The
system was initialised using preprogramed
basal rates, total daily insulin doses, and the
participants’ weights.
In all, 16 women completed all study-
related procedures (mean age, 34 ± 4 years;
A1c, 6.8 ± 0.6%; duration of diabetes, 23.6
± 7 years; week of gestation at randomisa-
tion, 14 ± 3.3 weeks). The primary outcome
was the percentage of time that overnight
glucose levels were within the target range
(63–140 mg/dL [3.50–7.77 mmol/L]). Use
of closed-loop insulin delivery increased
the time in target by 15% (closed-loop
74.7% vs SAP 59.5%; P = 0.002). Average
glucose was lower with closed-loop control
both overnight and over a 24-hour period
(overnight, P = 0.009; 24-hour period, P <
0.001). There was no difference in the time
Improved glycaemic control
with closed-loop insulin in
pregnant diabetic women
By Jennifer Sherr, MD, PhD
Advertisers
Novo Nordisk
Levemir/NovoRapid 7
NovoMix30 9
Just as closed-loop insulin delivery studies have
progressed from being performed solely in supervised,
inpatient research centres to home environments with
no remote monitoring, the population being studied
has expanded. Recently, these systems have been
assessed in those with type 1 diabetes deemed as
vulnerable populations – including young children1,2
and pregnant women.3,4
Upcoming
Endocrinology
conferences 2016
OCTOBER
13–15 October | Seoul, South Korea
International Conference on Diabetes and
Metabolism 2016
19–22 October | Milan, Italy
17th Congress of the European
Neuroendocrine Association 2016
31 October–4 November | New Orleans, USA
The Obesity Society: Annual Scientific Meeting
NOVEMBER
7–9 November | Brighton, UK
Society for Endocrinology BES 2016
meetings/2016/sfebes2016
11–12 November | Vienna, Austria
Prevention Models of Obesity and
Cardiovascular Diseases International
Symposium 2016
DECEMBER
1–3 December | Universal City, California, USA
14th Annual World Congress on Insulin
Resistance, Diabetes, and Cardiovascular
Disease 2016
2–4 December | Atlanta, USA
9th World Congress on Prevention of Diabetes
and its Complications 2016
FEATURE ARTICLE
VOL. 1 • No. 3 • 2016
15