The interaction between glycemic control and sleep characteristics in patients with type 1 Diabetes Mellitus

Type 1 diabetes mellitus (DM1) is caused by destruction of pancreatic β-cells
which results in absolute insulin deficiency. Intensive insulin therapy is
essential for optimal glucoregulation, because diabetic complications are
determined by the degree of long term hyperglycemia [1]. However,
glucoregulation can not be normalized in patients with DM1 despite intensive
insulin therapy and/or lifestyle adaption. This is reflected in relatively
large variations in blood glucose levels and relatively high HbA1c levels
compared to healthy subjects. [2]

Sleep duration and sleep quality are determinants of glucose tolerance:

Normal glucose regulation has a diurnal pattern with variations in glucose
tolerance, in which sleep plays a key role. [3] There is a strong relation
between sleep duration and glucoregulation. Partial sleep restriction during
one or multiple nights induces glucose intolerance and a reduction in the acute
insulin response to glucose in healthy volunteers. [4] In accordance, we
demonstrated that reduction of sleep duration even for a single night reduced
insulin sensitivity by ~20% in both healthy subjects and patients with type 1
Diabetes Mellitus. [5,6]
In addition to sleep duration, sleep quality is another determinant of
glucoregulation. Selective suppression of slow wave sleep (SWS) in healthy
subjects, without any change in sleep duration, induces marked glucose
intolerance measured by i.v. tolerance test (ivGTT), implying that interference
with sleep quality impairs subsequent glucoregulation, probably secondary to
increased sympathetic nervous activity. [7] The effects of controlled
interference with sleep quality on insulin sensitivity measured by the
euglycemic clamp method has not been performed in healthy subjects and patients
with DM1.

Metabolic regulation affects sleep quality:

Patients with DM have alterations in sleep characteristics. [8,9] Lamond et al.
[10] demonstrated that one-third of the patients with DM report sleep
difficulties. Other studies showed that patients with DM2 have more difficulty
initiating and maintaining sleep and more excessive daytime sleepiness compared
with healthy controls. [11-15]

There are only a few studies on sleep patterns in patients with DM1.
Jauch-Chara et al. [16] provided evidence for altered sleep architecture in
patients with DM1 under non-hypoglycaemic conditions. These patients showed a
tendency toward less deep sleep, with decreased amounts of slow wave sleep
(SWS) during the first half of the night.

Conversely, we hypothesize that hyperglycemia in DM1 may also affect sleep
characteristics. If this appears to be true, this may induce a vicious circle:
metabolic dysregulation impairs sleep characteristics, and, in turn, impaired
sleep quality/duration results in insulin resistance and impaired glucose
intolerance.

This study aims to investigate the hypothesis :
• Hyperglycemia alters sleep characteristics (i.e. sleep duration and/or
quality) in patients with DM1.
• Impaired sleep quality decreases insulin-mediated glucoregulation in healthy
subjects and in patients with DM1.

•To determine the effect of hyperglycemic dysregulation on sleep
characteristics (duration and quality), in patients with DM1
'What are the effects of controlled hyperglycemia on sleep characteristics
(sleep duration and/or quality) in patients with DM1?'

•Conversely, to examine the effect of decreased sleep quality on basal and
insulin-mediated glucoregulation in healthy subjects and in patients with DM1
' What is the effect of selective suppression of slow wave sleep (SWS) on basal
and insulin-mediated glucoregulation in healthy controls and in patients with
DM1?'

Prospective intervention study

Patients with DM1 will participate 4 nights:

•First night: normal sleep with PSG in order to become accustomed to the sleep
studies and to exclude patients with sleep disorders
•Normoglycemic night: normoglycemia and normal sleep with PSG to characterize
normal sleep characteristics and basal and insulin mediated glucoregulation
measured with the hyperinsulinemic euglycemic clamp technique.
•Hyperglycemic night: 50% reduction in basal and bolus insulin infusions to
induce a hyperglycemia to assess the effect of hyperglycemia on the sleep
characteristics, measured with PSG
•Night with impaired sleep quality: selective suppression of the slow-wave
sleep and assess the effect on basal and insulin-mediated glucoregulation
measured with the hyperinsulinemic euglycemic clamp technique.

The sequences of nights 2, 3, and 4 will be determined by balanced assignment

The healthy subjects will participate 3 nights:
•First night: normal night with PSG in order to become accustomed to the sleep
studies and to exclude subjects with sleep disorders
•Control night: normoglycemia and normal sleep to characterize normal sleep
characteristics and the next day subsequent basal and insulin-mediated
glucoregulation measured with the hyperinsulinemic euglycemic clamp technique.
•Night with impaired sleep quality: selective suppression of the slow-wave
sleep and assess the effect on basal and insulin mediated glucoregulation
measured with the hyperglycemic euglycemic clamp technique.

The sequences of nights 2 and 3 will be determined by balanced assignment

normoglycemic vs. hyperglycemic condition

normal sleep vs. selective suppression of the slow-wave sleep

Burden:

Subjects will sleep 3 (healthy controls) or 4 nights (diabetic patients) in our
research centre with polysomnography; of which 2 nights of normal sleep, 1
night with hyperglycemia (only in diabetic patients), 1 night with selective
suppression of the slow wave sleep. No adverse events are aspects from this.
Subsequently, the subjects will be in our hospital (after studyday 2, and 4)
for 2 whole days. During these days, subjects will lie in bed and blood will be
drawn from an infusion.
- Basal experiment: infusion of labeled glucose which is not
radioactive en therefore no adverse effects are expected.
- Hyperinsulinemic euglycaemic clamp: infusion of insulin and labeled
glucose. Blood glucose measurements will be made at regular
time intervals to adjust for glucose infusion and prevent
hypoglycemia.

hyperglycemic night: 50% reduction in basal and bolus insulin infusion in order
to get and maintain a hyperglycemia with glucose levels between 15 and 20
mmol/L. The blood glucose levels will be carefully evaluated by the research
physician during the night. No adverse events are aspects from this.

Halvering van de basale en bolus bolussen van de onderhuidse insuline pomp,
waardoor er een hyperglycemie wordt bereikt. Aangezien we de patient goed
instrueren om klachten van een hyperglycemie te herkennen, verwachten we geen
bijwerkingen. Ook houden we de patient gedurende de gehele nacht goed in de
gaten.