Continuous Glucose Monitoring · Continuous Glucose Monitoring IN CONTROL of Type 1 Diabetes...

Continuous Glucose

MonitoringIN CONTROL of Type 1 Diabetes

Koen van Beers

Continuous Glucose Monitoring

IN CONTROL of Type 1 Diabetes

Cornelis A.J. van Beers

CONTINUOUS GLUCOSE MONITORING

IN CONTROL OF TYPE 1 DIABETES

ISBN nummer 978-94-6233-839-5Author Cornelis A.J. van BeersCover Margreet Schekkerman by Marieke WisseLayout Marieke Wisse and Annelies Wisse, AmsterdamPrinted by Gildeprint Drukkerijen, Enschede

© C.A.J. van Beers, Amsterdam, The Netherlands, 2017No part of this thesis may be reproduced, stored or transmitted in any form or by any means, without prior permission of the author.

Financial support by Eli Lilly, Sanofi and Medtronic for the work within thisthesis is gratefully acknowledged

The printing of the thesis was additionally kindly supported by: Boehringer Ingelheim bv and ChipSoft.

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad Doctor aan

de Vrije Universiteit Amsterdam,

op gezag van de rector magnificus

prof.dr. V. Subramaniam,

in het openbaar te verdedigen

ten overstaan van de promotiecommissie

van de Faculteit der Geneeskunde

op maandag 15 januari 2018 om 13.45 uur

in de aula van de universiteit,

De Boelelaan 1105

VRIJE UNIVERSITEIT


IN CONTROL of Type 1 Diabetes

door

Cornelis Antonius Johannes van Beers

geboren te Roosendaal

promotoren:

copromotoren:

prof.dr. M.H.H. Kramer

prof.dr. F.J. Snoek

dr. E.H. Serné

dr. P.H.L.M. Geelhoed-Duijvestijn

leescommissieprof.dr. P.T.A. Lips

prof.dr. M. Nieuwdorp

dr. R.G. Ijzerman

prof.dr. B.M. Frier

prof.dr. R.J. Heine

prof.dr. F. Pouwer

dr. H.W. de Valk

Aan mijn ouders

Contents

General introduction and outline of the thesis

Continuous Glucose Monitoring:

Impact on Hypoglycaemia

Design and rationale of the IN CONTROL trial: the effects of

real-time continuous glucose monitoring on glycaemia and quality

of life in patients with type 1 diabetes mellitus and impaired

awareness of hypoglycaemia

Continuous glucose monitoring for type 1 diabetes patients

with impaired awareness of hypoglycaemia (IN CONTROL): a

randomised, open-label, crossover trial

Continuous glucose monitoring in patients with type 1 diabetes

and impaired awareness of hypoglycaemia: also effective in

patients with psychological distress?

Keeping safe. Continuous glucose monitoring in persons with type

1 diabetes and impaired awareness of hypoglycaemia:

a qualitative study

The relation between HbA1c and hypoglycaemia revisited; a

secondary analysis from an intervention trial in patients with type

1 diabetes and impaired awareness of hypoglycaemia

Chapter 1

Chapter 2

Chapter 3

Chapter 4

Chapter 5

Chapter 6

Chapter 7

13

33

55

79

107

121

139

Summary & general discussion

Nederlandse samenvatting Summary in Dutch

Author affiliations

Dankwoord Acknowledgements

Biografie Biography

Chapter 8

Chapter 9

155

171

180

182

188

1

General introduction

and outline of the thesis

Ch

apte

r 1

14

TYPE 1 DIABETES

Type 1 diabetes (T1D) constitutes roughly 5% to 10% of the total cases of diabetes.1

Worldwide, the incidence and prevalence of T1D vary significantly.2 In the Netherlands,

the incidence rate is approximately 15 cases per 100.000 persons per year.3 Importantly,

the EURODIAB registers showed that the incidence of T1D increases annually at an

alarming rate ranging from 0.6% to 9.3%,4 especially at younger age, and the incidence

of T1D in children <5years of age will be doubled from 2009 to 2020 if this is maintained.

Type 1 diabetes is thought to occur as a result of an immune mediated or associated

destruction of insulin-producing pancreatic β cells.5 Destruction of these β cells

results in absolute insulin deficiency necessitating lifelong treatment with exogenous

insulin. Since the Diabetes Control and Complications Trial (DCCT) demonstrated that

strict glycaemic control (a lower HbA1c value) significantly lowers the risk of both

microvascular6 and macrovascular7 complications in patients with T1D, intensive

insulin treatment has become the standard. Intensive insulin therapy consists of multiple

daily injections (MDI) of long and short actin insulins or continuous subcutaneous

insulin infusion (CSII) by insulin pumps. Self-management of diabetes with CSII or MDI

requires patients with T1D to frequently self-monitor their glucose values, either by

finger sticks, or, since 2006, by continuous glucose monitoring. Although glycaemic

outcomes have overall significantly improved in the past decades, achievement of

constant normoglycaemia is an elusive goal for the majority of T1D patients.8 Striving

for normoglycaemia is associated with an increased risk of hypoglycaemia.9 Indeed,

despite advances in diabetes treatment , hypoglycaemia remains the main side-effect

of insulin therapy and barrier to reaching sustained glycaemic control.10

15

HYPOGLYCAEMIA

Definition and epidemiology

The biochemical cut-off for hypoglycaemia is a matter of continuous debate.11

The American Diabetes Association (ADA) proposed a biochemical definition of

hypoglycaemia as a plasma glucose of ≤3.9 mmol/L, because, in healthy individuals,

the stimulation of glucagon and adrenaline occurs around a plasma glucose level of

3.9 mmol/L (Figure 1).12,13 Also, in healthy individuals, an antecedent hypoglycaemic

episode of 3.9 mmol/L can cause suppression of subsequent adrenaline, glucagon and

muscle sympathetic nerve activity response to another hypoglycaemic episode on the

second day.14 However, many clinicians believe this relatively high cut-off value causes

3.8 mmol/LIncreased glucagon

secretionIncreased adrenaline

secretion

2.8 mmol/LCognitive dysfunction

4.6 mmol/LInhibition endogenous insulin secretion

3.2 - 2.8 mmol/LOnset of hypoglycaemic symptoms

<1.5 mmol/LSevere neuroglycopenia (e.g. coma, seizures, brain death)

Plasma glucose level (mmol/L)

Figure 1. Hierarchy of responses to falling arterial plasma glucose concentrations. (Adapted from Cryer64)

-5-

-4-

-3-

-2-

-1-

-0-

Ch

apte

r 1

16

an increase in the incidence of clinically meaningless biochemical hypoglycaemia and

increase in the prevalence of impaired awareness of hypoglycaemia. Hypoglycaemia

is usually clinically categorized by whether a patient is able to self-treat or requires

assistance of others. Mild hypoglycaemia is usually defined as an episode in which a

person is able to recognize and self-treat a low level of blood glucose, whereas severe

hypoglycaemia is often defined as a hypoglycaemic event requiring assistance of a

third party.12

In patients with T1D, the mean incidence of mild hypoglycaemia is roughly one to two

events per patient per week and the incidence of severe hypoglycaemia is approximately

0.2 to 3.2 events per patient per year.15,16 The risk of severe hypoglycaemia increases

with increasing duration of T1D. In patients with a long disease duration (>15 years),

a prevalence of up to 46%, and a mean rate of 3.2 episodes per subject-year have been

reported.17 Severe hypoglycaemia is known to cluster in a subgroup of the population;

only a small proportion experience multiple episodes and many experience none.18

Consequences of hypoglycaemia

Hypoglycaemia is not benign, but has important physical and psychosocial

consequences.15,19 Hypoglycaemia interferes with many aspects of daily life,

including sleep, driving, exercise, social functioning and employment.19 In patients

with type 2 diabetes with significant cardiovascular risk, hypoglycaemia probably

increases the risk of cardiovascular events,20-22 although causality remains difficult to

prove.23 Furthermore, hypoglycaemia impairs cerebral function and might promote

permanent cognitive decline.24,25 Hypoglycaemia can also have a profound effect on

psychosocial well-being and causes fear of hypoglycaemia.26-28 Also, healthcare costs

are substantially increased because of hypoglycaemia.29 Importantly, hypoglycaemia

can be fatal, with mortality estimates ranging from 4 to 10 percent of deaths in T1D

patients diagnosed in childhood or early adulthood and dying before the age of 40

years.30,31 A registry-based observational study showed that in T1D patients younger

than 30 years, 31.4% of deaths was caused by diabetic ketoacidosis or hypoglycaemia.32

Normal counter-regulation and symptomatology

The brain is almost totally dependent on carbohydrate as a fuel and since it cannot

store or synthesize glucose, depends on a continuous supply from the blood.33 Although

recent neuroimaging techniques have revealed that recurrent hypoglycaemia causes

cerebral adaptations,34 the potentially serious effects of hypoglycaemia on cerebral

function mean that not only are stable blood glucose concentrations maintained

17

under physiological conditions, but if hypoglycaemia occurs, counter-regulatory

mechanisms are initiated to combat it (Figure 1 and Figure 2).35,36 The counter-regu-

latory mechanisms are preceded by suppression of endogenous insulin secretion.

In case of acute hypoglycaemia, glucagon and adrenaline are the most important

counter-regulatory hormones, increasing glycogenolysis and stimulate gluconeo-

genesis. In addition, adrenaline reduces glucose utilization peripherally and inhibits

insulin secretion. Cortisol and growth hormone counteract prolonged hypoglycaemia

by increasing gluconeogenesis and reducing glucose utilization. Also, the autonomic

nervous system (both sympathetic and parasympathetic components) is activated

during hypoglycaemia and is responsible for many of the physiological changes

Decrements in insulin and increments in glucagon are lost and increments in epinephrine and neurogenic symptoms are often

attenuated in type 1diabetes. SNS: sympathetic nervous system; PNS: parasympathetic nervous system; NE norepinephrine;

Ach: acetylcholine; α cell: pancreatic islet α cells; β cell: pancreatic islet β cells. (Adapted from Cryer33)

Figure 2. Physiological and behavioural defences against hypoglycaemia

Peripheral sensors

Decreased glucose

Decreased insulin

Decreased insulin

Increased glucagon

Increased adrenaline

Decreased glucose

clearance

(Often attenueted in T1DM)(Often attenueted in T1DM)

Increased glycogenolysis&

Increased gluconeogenesis

Increasedglucose production

Increasedlactate, aminoacids, glycerol

Increasedglucose

Increasedneurogenic symptoms

Increased ACh (sweating, hunger)

Increased NE (palpitations, tremor, arousal)

Increasedsympathoadrenal outflow

Muscle Kidney Fat

(SNS)

(PNS)

(Lost in T1DM) (Lost in T1DM)

CNS

Increasedingestion of carbohydrates

Ch

apte

r 1

18

and autonomic symptoms during hypoglycaemia (i.e. hunger, sweating, tremor,

palpitation). These autonomic symptoms are the tangible effects of sympathoadrenal

stimulation of end-organs such as the heart, sweat glands and muscle. The intensity of

the symptoms is heightened by the secretion of adrenaline, but the counter-regulatory

hormones are not critical to the generation of symptoms. Rather it is the activation of

central autonomic centres within the brain that generates the autonomic symptoms.

Other symptoms of hypoglycaemia, such as confusion, drowsiness, odd behaviour,

and difficulty speaking are termed neuroglycopenic and occur as a consequence

of cerebral glucose deprivation. Symptoms of hypoglycaemia, both autonomic and

neuroglycopenic,37 help to warn the individual that their blood glucose is falling low,

thereby encouraging the ingestion of carbohydrate, helping to restore glucose concen-

trations in addition to counter-regulation.

Counter-regulatory deficiencies and hypoglycaemia acquired syndromes

In patients with T1D, multiple physiologic defences against the development of

hypoglycaemia – decrements in insulin and increments in glucagon and epinephrine –

become comprised (Figure 2).38 Therapeutic insulin levels do not fall, and the glucagon

response to hypoglycaemia rapidly declines in patients with T1D.39 In addition, the

glycaemic threshold for the adrenaline response is often shifted towards lower plasma

glucose concentrations. The combination of an absent glucagon response and an

attenuated epinephrine response causes the clinical syndrome of defective counter-re-

gulation.33 In addition to defective counter-regulation, the sympathoadrenal activation

responsible for the generation of autonomic symptoms also becomes attenuated

with time. As a result of the lower threshold for sympathoadrenal activation, the

autonomic warning symptoms are partly or completely lost, the intensity of the

symptoms diminished, or present too late to elicit autonomic symptoms, and patients

fail to recognize them due to neuroglycopenia, thereby compromising the behavioural

defences against hypoglycaemia (ingestion of carbohydrates). This cascade of

events constitutes and reinforces the clinical syndrome of impaired awareness of

hypoglycaemia (IAH). Recurrent hypoglycaemia is thought to cause both counter-re-

gulatory failure and impaired awareness of hypoglycaemia. The combination of both

syndromes is called ‘Hypoglycaemia Associated Autonomic Failure’ (HAAF) and both

syndromes share a similar pathogenesis (Figure 3).33

Impaired awareness of hypoglycaemia

Impaired awareness of hypoglycaemia is clinically often defined by the loss of

the ability to perceive the onset of acute hypoglycaemia, but may also be manifested

19by a reduced intensity and/or number of symptoms, a change in symptom profile,

or a failure of the patient to interpret symptoms. Because IAH may mean different

things to different people, no international consensus exists regarding a practical

definition of IAH. Impaired awareness of hypoglycaemia is a preferable terminology

to ‘hypoglycaemia unawareness’, since almost no patients have complete loss of

Type 1 diabetes (no decrease in insulin, no increase in glucagon, therapeutic hyperinsulinemia)

Hypoglycaemia

Attenuated sympathoadrenal re-sponses to hypoglycaemia (HAAF)

Decreased adrenomedullary adrenaline responses

Decreased sympathic neural responses

Defective glucose counter-regulation

Impaired awareness of hypoglycaemia

Recurrent hypoglycaemia

Figure 3. Pathophysiology of Hypoglycaemia Associated Autonomic Failure (HAAF). (Adapted from Cryer33)

Ch

apte

r 1

20

hypoglycaemia-related symptoms. Impaired awareness of hypoglycaemia occurs in

roughly 25% of patients with longstanding T1D and renders them at a significantly

increased risk of severe hypoglycaemia.40,41 The most important factors associated

with IAH include increasing age, increasing duration of diabetes, and strict glycaemic

control.18,41 In clinical practice, IAH can best be assessed by using a clinical history.

In addition, for the purpose of guiding clinical practice and performing clinical

trials, multiple self-rating questionnaires have been developed,40,42 which show good

concordance with objective measures in adult patients with T1D.40,43 The Gold method

consists of one question: “do you know when hypoglycaemia is commencing?”. The

state of hypoglycaemia awareness is then assessed by using a 7-point visual analogue

scale, with 1 representing “always aware” and 7 representing “never aware”. A score

of ≥4 suggests IAH.40 Impaired awareness of hypoglycaemia has previously also been

assessed by use of clamp studies,38 but, obviously, this experimental setting barely

reflects real-life conditions. As antecedent hypoglycaemia has an important role in

the pathogenesis of IAH, rigorous avoidance of hypoglycaemia seems to be crucial for

restoring IAH,44 although this is very difficult to achieve. Various treatment strategies

have been proposed for T1D patients with IAH or problematic hypoglycaemia, including

structured diabetes education programs and supportive diabetes technologies, such as

CSII and continuous glucose monitoring (CGM).45

21

GLUCOSE MONITORING

Since its introduction in the early 1920s, clinicians have recognized that insulin

therapy goes hand in hand with glycaemic excursions, endorsing the need for glucose

monitoring.46 During the first half of the 20th century patients’ glycaemic control was

evaluated by means of urinary tests, first using copper reagent tablets and later glucose

oxidase impregnated dipsticks for semi-quantitative assessment of glycosuria.47,48 It

was well recognised then that urine testing had numerous limitations for diabetes

monitoring. First, fluid intake and urine concentration affected test results according

to the sensitivity of the reagent strip. Second, urine glucose could only be retrospective

of the current glycaemic status. Third, positive results only occurred when the

renal threshold for glucose was exceeded and this varied in longstanding diabetes

or pregnancy. Fourth, negative results did not distinguish between hypoglycaemia,

normoglycaemia and even mild hyperglycaemia.49 Fifth, correlation between urine

and plasma glucose has been shown to be inconsistent.50 Consequently, blood became

the preferred sample, most easily collected by fingertip capillary puncture, which

reflects ‘real-time’ blood glucose concentrations.

During the late 1960s and 1970s, the first test strips for measuring blood glucose

were developed, first for use in doctors’ offices, and during the 1970s the concept of

self-monitoring of blood glucose (SMBG) was more and more considered. These test

strips contained glucose oxidase, causing a biochemical reaction with glucose with

hydrogen peroxidase as the end product. The amount of hydrogen peroxide produced

in the glucose oxidase reaction was linked to an intensity of colour. By comparing the

colour with a standardized series of printed colours, the blood glucose level could

be estimated. The major limitation to this approach was influenced by the patients’

ability to perceive colour accurately. This problem was solved a couple of years later by

making an electronic measurement of the intensity of colour on the strip.48

The 1980s was an active phase in the evolution of glucose meters, which were

becoming easier to use, smaller in size, with more variation in design, often with

software memory to store and retrieve results. Reagent strips were also changing to

accept smaller volumes of blood, and some were barcoded for auto calibration and

quality assurance.

Most significantly, towards the end of the 1980s, the first enzyme electrode strips

were introduced, providing a choice of instrument (i.e., using either reflectance

or electrochemical principles) to measure blood glucose. The mechanisms of these

novel blood glucose meters was no longer based on a photometric approach, but on

Ch

apte

r 1

22

an electrochemical reaction, with glucose generating an electrical current related to

the glucose concentration.48 These blood glucose meters allowed patients to check

their glycaemic status at home. To date, the electrochemical blood glucose meters are

still being used for glucose monitoring and insulin dosing. The amount of mealtime

insulin supplemented is co-determined by SMBG values. Patients have to monitor

their blood glucose multiple times daily to determine the amount of insulin necessary

for mealtime bolus, to evaluate if the bolus was correct, but also in other daily

circumstances, such as before driving, before and after exercise, or when patients

feel hyper- or hypoglycaemia. Although the number of SMBG measurements per day

is tightly correlated with a lower HbA1c,51 SMBG has many shortcomings (e.g. it is a

hassle and results in inconveniences and incompleteness of data). As a consequence,

even when patients perform more than the advised number of SMBG measurements

per day, most patients with T1D do not reach and sustain constant normoglycaemia.8


Although the technique of continuous glucose monitoring became available during

the late 1990s, it was not until 2006 that real-time continuous glucose monitoring (CGM)

was introduced to assist patients in their self-management.52 Present CGM systems

that are available use small minimally invasive sensors which measure interstitial

glucose levels via the glucose-oxidase reaction and translate this into glucose values

by means of calibrations.53,54 The CGM systems provide this information every five or

ten minutes, with a delay of approximately 5 to 15 minutes. The added value lies in

the semi-continuous display of ‘current’ glucose values, visualization of glucose trends

and the availability of alarms that can be set to warn for impending hypoglycaemia

or hyperglycaemia.55 First generation CGM systems were used as stand-alone devices.

Next generation CGM systems are connected to insulin pumps (sensor-augmented

pump therapy; SAPT), but do not interfere with insulin delivery automatically. These

CGM systems therefore only act as behaviour modifiers, rather than insulin dose

adjustment tools. The newest generation SAPT systems however have a (predicted)

low-glucose suspend (LGS) feature, which automatically interrupts insulin adminis-

tration when glucose falls below a pre-set threshold.56,57 Recently, the Food and Drug

Administration (FDA) approved the first hybrid closed-loop insulin delivery system,

combining user-delivered pre-meal boluses with automated inter-prandial insulin

delivery.58

Continuous glucose monitoring (CGM) has shown to reduce HbA1c without

increasing hypoglycaemia, with the largest effect seen in patients with the highest

HbA1c at baseline.59 CGM may also help to reduce the adverse psychosocial effects

23

of (unpredictable recurrent hypoglycaemia in) T1D, but evidence is limited and

inconsistent.60-62 However, whether CGM improves glycaemic control and quality

of life more than self-monitoring of blood glucose (SMBG) in patients with T1D and

IAH has yet to be determined.63-66 It is important to note that CGM can aid patients to

self-manage their diabetes more precisely, provided they are capable of handling the

device and data feedback adequately, with support of a diabetes health care team.67,68 In

this context, the question comes up whether CGM is suitable and beneficial for patients

with an unfavourable psychological profile, e.g. with high psychological distress.

Psychological distress is common in diabetes, including low emotional well-being, high

diabetes-related distress, and fear of hypoglycaemia, negatively affecting patients’

daily self-care and glycaemic control.69-73 With CGM, patients are faced with real-time

feedback on blood glucose variation and alarms that may be experienced as stressful

and difficult to handle for those with pre-existing high levels of distress.67,74,75 To date, no

trials investigating the effect of CGM have studied the modifying role of psychological

distress on treatment outcomes in this patient group. Also, CGM experiences in type1

diabetes patients with problematic hypoglycaemia have not been explored in-depth.

Understanding patients’ expectations and perceived benefits and losses of CGM use

could explain, at least in part, the observed between-patient differences in adherence

to CGM and effectiveness. Furthermore, insights into the expectations and experiences

of CGM could offer guidance for clinicians and researchers to address these factors,

which might improve adherence to CGM and effectiveness of CGM. We hypothesize

that CGM is a valuable tool in the (safe) treatment of adult patients with T1D and IAH.

Ch

apte

r 1

24

OUTLINE OF THE THESIS

The aim of this thesis was to investigate the effect of CGM on glycaemic control and

psychological outcomes in patients with T1D and impaired awareness of hypoglycaemia.

In Chapter 2 of this thesis, we provide an overview of the previously published CGM

trials in patients with T1D and mainly focus on the hypoglycaemia-related outcomes.

Sensor-augmented pump therapy with (predictive) low-glucose suspension is discussed

separately. Also, trials performed in patients T1D and IAH are discussed. Chapter 3

describes the design and rationale of our randomised, cross-over trial investigating

the effect of CGM on glycaemic control and psychological outcomes in patients with

T1D and IAH. Subsequently, in Chapter 4, we tested the hypothesis that CGM improves

glycaemic control and psychological outcomes, and prevents severe hypoglycaemia

in adult patients with T1D and IAH, by comparing the glycaemic and psychological

outcomes between a period of glucose monitoring by CGM (intervention) and a period of

standard glucose monitoring by SMBG (control). In Chapter 5, we investigate whether

psychological distress modifies the effect of CGM on glycaemic control in patients with

T1D and IAH. In Chapter 6, we conduct a supplementary qualitative study, based on

semi-structured interviews, to further our understanding of the use, perceptions and

experiences of CGM in this typical population at high risk of hypoglycaemia. In Chapter

7, we re-assess the previously shown but recently disputed association between HbA1c

and severe hypoglycaemia in patients with T1D and IAH. Finally, in Chapter 8, we

discuss our data and suggest implications for clinical practice.

25

1. American Diabetes Association.

Diagnosis and classification of

diabetes mellitus. Diabetes Care

2009; 32 Suppl 1: S62-S7.

2. Maahs DM, West NA, Lawrence JM,

Mayer-Davis EJ. Epidemiology of

type 1 diabetes. Endocrinol Metab

Clin North Am 2010; 39(3): 481-97.

3. EURODIAB ACE Study Group.

Variation and trends in incidence of

childhood diabetes in Europe. Lancet

2000; 355(9207): 873-6.

4. Patterson CC, Dahlquist GG, Gyurus

E, Green A, Soltesz G. Incidence

trends for childhood type 1 diabetes

in Europe during 1989-2003 and

predicted new cases 2005-20: a

multicentre prospective registration

study. Lancet 2009; 373(9680):

2027-33.

5. Bluestone JA, Herold K, Eisenbarth G.

Genetics, pathogenesis and clinical

interventions in type 1 diabetes.

Nature 2010; 464(7293): 1293-300.

6. The Diabetes Control and

Complications Trial Research Group.

The effect of intensive treatment

of diabetes on the development

and progression of long-term

complications in insulin-dependent

diabetes mellitus. N Engl J Med 1993;

329(14): 977-86.

7. Nathan DM, Cleary PA, Backlund JY,

et al. Intensive diabetes treatment

and cardiovascular disease in

patients with type 1 diabetes. N Engl

J Med 2005; 353(25): 2643-53.

8. Miller KM, Foster NC, Beck RW, et

al. Current state of type 1 diabetes

treatment in the U.S.: updated

data from the T1D Exchange clinic

registry. Diabetes Care 2015; 38(6):

971-8.



Hypoglycemia in the Diabetes

Control and Complications Trial.

Diabetes 1997; 46(2): 271-86.

10. Cryer PE. Hypoglycaemia: the

limiting factor in the glycaemic

management of Type I and Type II

diabetes. Diabetologia 2002; 45(7):

937-48.

11. Frier BM. Defining hypoglycaemia:

what level has clinical relevance?

Diabetologia 2009; 52(1): 31-4.

12. American Diabetes Association

Workgroup on Hypoglycemia.

Defining and reporting hypoglycemia

in diabetes: a report from the

American Diabetes Association


Diabetes Care 2005; 28(5): 1245-9.

13. Schwartz NS, Clutter WE, Shah

SD, Cryer PE. Glycemic thresholds

for activation of glucose

counterregulatory systems are

higher than the threshold for

REFERENCES

Ch

apte

r 1

26

symptoms. J Clin Invest 1987; 79(3):

777-81.

14. Davis SN, Shavers C, Mosqueda-

Garcia R, Costa F. Effects of differing

antecedent hypoglycemia on

subsequent counterregulation in

normal humans. Diabetes 1997;

46(8): 1328-35.

15. Frier BM. Hypoglycaemia in diabetes

mellitus: epidemiology and clinical

implications. Nat Rev Endocrinol

2014; 10(12): 711-22.

16. Ostenson CG, Geelhoed-Duijvestijn P,

Lahtela J, Weitgasser R, Markert JM,

Pedersen-Bjergaard U. Self-reported

non-severe hypoglycaemic events

in Europe. Diabet Med 2014; 31(1):

92-101.

17. UK Hypoglycaemia Study Group.

Risk of hypoglycaemia in types 1

and 2 diabetes: effects of treatment

modalities and their duration.

Diabetologia 2007; 50(6): 1140-7.

18. Pedersen-Bjergaard U, Pramming

S, Heller SR, et al. Severe

hypoglycaemia in 1076 adult patients

with type 1 diabetes: influence of

risk markers and selection. Diabetes

Metab Res Rev 2004; 20(6): 479-86.

19. Frier BM. How hypoglycaemia

can affect the life of a person with

diabetes. Diabetes Metab Res Rev

2008; 24(2): 87-92.

20. Duckworth W, Abraira C, Moritz T,

et al. Glucose control and vascular

complications in veterans with type

2 diabetes. N Engl J Med 2009; 360(2):

129-39.

21. Gerstein HC, Miller ME, Byington

RP, et al. Effects of intensive glucose

lowering in type 2 diabetes. N Engl J

Med 2008; 358(24): 2545-59.

22. Patel A, MacMahon S, Chalmers J, et

al. Intensive blood glucose control

and vascular outcomes in patients

with type 2 diabetes. N Engl J Med

2008; 358(24): 2560-72.

23. Bonds DE, Miller ME, Bergenstal

RM, et al. The association between

symptomatic, severe hypoglycaemia

and mortality in type 2 diabetes:

retrospective epidemiological

analysis of the ACCORD study. BMJ

2010; 340: b4909.

24. Feinkohl I, Aung PP, Keller M, et al.

Severe hypoglycemia and cognitive

decline in older people with type

2 diabetes: the Edinburgh type 2

diabetes study. Diabetes Care 2014;

37(2): 507-15.

25. Warren RE, Frier BM.

Hypoglycaemia and cognitive

function. Diabetes Obes Metab 2005;

7(5): 493-503.

26. Gonder-Frederick L, Nyer M,

Shepard JA, Vajda K, Clarke W.

Assessing fear of hypoglycemia in

children with Type 1 diabetes and

their parents. Diabetes Manag (Lond)

2011; 1(6): 627-39.

27. Gonder-Frederick LA, Cox DJ, Bobbitt

SA, Pennebaker JW. Mood changes

associated with blood glucose

fluctuations in insulin-dependent

27

diabetes mellitus. Health Psychol

1989; 8(1): 45-59.

28. Gonder-Frederick LA, Clarke WL,

Cox DJ. The Emotional, Social, and

Behavioral Implications of Insulin-

Induced Hypoglycemia. Semin Clin

Neuropsychiatry 1997; 2(1): 57-65.

29. Foos V, Varol N, Curtis BH, et al.

Economic impact of severe and

non-severe hypoglycemia in patients

with Type 1 and Type 2 diabetes in

the United States. J Med Econ 2015;

18(6): 420-32.

30. Feltbower RG, Bodansky HJ,

Patterson CC, et al. Acute

complications and drug misuse

are important causes of death for

children and young adults with type

1 diabetes: results from the Yorkshire

Register of diabetes in children and

young adults. Diabetes Care 2008;

31(5): 922-6.

31. Patterson CC, Dahlquist G, Harjutsalo

V, et al. Early mortality in EURODIAB

population-based cohorts of type

1 diabetes diagnosed in childhood

since 1989. Diabetologia 2007; 50(12):

2439-42.

32. Lind M, Svensson AM, Kosiborod M,

et al. Glycemic control and excess

mortality in type 1 diabetes. N Engl J

Med 2014; 371(21): 1972-82.

33. Cryer PE. Diverse causes of

hypoglycemia-associated autonomic

failure in diabetes. N Engl J Med

2004; 350(22): 2272-9.

34. Rooijackers HM, Wiegers EC, Tack

CJ, van der Graaf M, de Galan BE.

Brain glucose metabolism during

hypoglycemia in type 1 diabetes:

insights from functional and

metabolic neuroimaging studies. Cell

Mol Life Sci 2016; 73(4): 705-22.

35. Cryer PE, Gerich JE. Glucose

counterregulation, hypoglycemia,

and intensive insulin therapy in


313(4): 232-41.

36. Cryer PE. Glucose counterregulation

in man. Diabetes 1981; 30(3): 261-4.

37. Cox DJ, Gonder-Frederick L, Antoun

B, Cryer PE, Clarke WL. Perceived

symptoms in the recognition of

hypoglycemia. Diabetes Care 1993;

16(2): 519-27.

38. Mokan M, Mitrakou A, Veneman T,

et al. Hypoglycemia unawareness in

IDDM. Diabetes Care 1994; 17(12):

1397-403.

39. Gerich JE, Langlois M, Noacco C,

Karam JH, Forsham PH. Lack of

glucagon response to hypoglycemia

in diabetes: evidence for an intrinsic

pancreatic alpha cell defect. Science

1973; 182(4108): 171-3.

40. Gold AE, MacLeod KM, Frier BM.

Frequency of severe hypoglycemia

in patients with type I diabetes

with impaired awareness of


17(7): 697-703.

41. Geddes J, Schopman JE, Zammitt NN,

Frier BM. Prevalence of impaired

awareness of hypoglycaemia in

Ch

apte

r 1

28

adults with Type 1 diabetes. Diabet

Med 2008; 25(4): 501-4.

42. Clarke WL, Cox DJ, Gonder-Frederick

LA, Julian D, Schlundt D, Polonsky W.

Reduced awareness of hypoglycemia

in adults with IDDM. A prospective

study of hypoglycemic frequency

and associated symptoms. Diabetes

Care 1995; 18(4): 517-22.

43. Geddes J, Wright RJ, Zammitt NN,

Deary IJ, Frier BM. An evaluation

of methods of assessing impaired

awareness of hypoglycemia in type 1

diabetes. Diabetes Care 2007; 30(7):

1868-70.

44. Cranston I, Lomas J, Maran A,

Macdonald I, Amiel SA. Restoration

of hypoglycaemia awareness in

patients with long-duration insulin-

dependent diabetes. Lancet 1994;

344(8918): 283-7.

45. Yeoh E, Choudhary P, Nwokolo M,

Ayis S, Amiel SA. Interventions

That Restore Awareness of

Hypoglycemia in Adults With Type

1 Diabetes: A Systematic Review and

Meta-analysis. Diabetes Care 2015;

38(8): 1592-609.

46. 46. Fletcher AA CW. The blood sugar

following insulin administration

and the symptom complex:

hypoglycemia. J Metab Res 1922; 2:

637-49.

47. Clarke SF, Foster JR. A history of

blood glucose meters and their

role in self-monitoring of diabetes

mellitus. Br J Biomed Sci 2012; 69(2):

83-93.

48. Clarke SF, Foster JR. A history of

blood glucose meters and their

role in self-monitoring of diabetes

mellitus. Br J Biomed Sci 2012; 69(2):

83-93.

49. Goldstein DE, Little RR, Lorenz RA,

Malone JI, Nathan DM, Peterson

CM. Tests of glycemia in diabetes.

Diabetes Care 2004; 27 Suppl 1:

S91-S3.

50. Hayford JT, Weydert JA, Thompson

RG. Validity of urine glucose

measurements for estimating plasma

glucose concentration. Diabetes Care

1983; 6(1): 40-4.

51. Evans JM, Newton RW, Ruta DA,

MacDonald TM, Stevenson RJ, Morris

AD. Frequency of blood glucose

monitoring in relation to glycaemic

control: observational study with

diabetes database. BMJ 1999;

319(7202): 83-6.

52. Garg S, Zisser H, Schwartz S, et al.

Improvement in glycemic excursions

with a transcutaneous, real-time

continuous glucose sensor: a

randomized controlled trial. Diabetes

Care 2006; 29(1): 44-50.

53. Feldman B, Brazg R, Schwartz S,

Weinstein R. A continuous glucose

sensor based on wired enzyme

technology -- results from a 3-day

trial in patients with type 1 diabetes.

Diabetes Technol Ther 2003; 5(5):

769-79.

54. Keenan DB, Mastrototaro JJ,

29

Voskanyan G, Steil GM. Delays in

minimally invasive continuous

glucose monitoring devices: a review

of current technology. J Diabetes Sci

Technol 2009; 3(5): 1207-14.

55. Bode B, Gross K, Rikalo N, et al.

Alarms based on real-time sensor

glucose values alert patients to hypo-

and hyperglycemia: the guardian

continuous monitoring system.


105-13.

56. Bergenstal RM, Klonoff DC, Garg SK,

et al. Threshold-based insulin-pump

interruption for reduction of

hypoglycemia. N Engl J Med 2013;

369(3): 224-32.

57. Maahs DM, Calhoun P, Buckingham

BA, et al. A randomized trial of a

home system to reduce nocturnal

hypoglycemia in type 1 diabetes.

Diabetes Care 2014; 37(7): 1885-91.

58. Bergenstal RM, Garg S, Weinzimer

SA, et al. Safety of a Hybrid

Closed-Loop Insulin Delivery System

in Patients With Type 1 Diabetes.

JAMA 2016; 316(13): 1407-8.

59. Pickup JC, Freeman SC, Sutton

AJ. Glycaemic control in type 1

diabetes during real time continuous

glucose monitoring compared with

self monitoring of blood glucose:

meta-analysis of randomised

controlled trials using individual

patient data. BMJ 2011; 343: d3805.

60. Hermanides J, Norgaard K,

Bruttomesso D, et al. Sensor-

augmented pump therapy lowers

HbA(1c) in suboptimally controlled

Type 1 diabetes; a randomized

controlled trial. Diabet Med 2011;

28(10): 1158-67.

61. Beck RW, Lawrence JM, Laffel L,

et al. Quality-of-life measures in

children and adults with type 1

diabetes: Juvenile Diabetes Research

Foundation Continuous Glucose

Monitoring randomized trial.

Diabetes Care 2010; 33(10): 2175-7.

62. Polonsky WH, Hessler D, Ruedy KJ,

Beck RW, Group DS. The Impact of

Continuous Glucose Monitoring on

Markers of Quality of Life in Adults

With Type 1 Diabetes: Further

Findings From the DIAMOND

Randomized Clinical Trial. Diabetes

Care 2017.

63. Choudhary P, Ramasamy S, Green

L, et al. Real-time continuous

glucose monitoring significantly

reduces severe hypoglycemia in

hypoglycemia-unaware patients with

type 1 diabetes. Diabetes Care 2013;

36(12): 4160-2.

64. Langendam M, Luijf YM, Hooft L,

DeVries JH, Mudde AH, Scholten

RJ. Continuous glucose monitoring

systems for type 1 diabetes mellitus.

Cochrane Database Syst Rev 2012; 1:

CD008101.

65. Little SA, Leelarathna L, Walkinshaw

E, et al. Recovery of hypoglycemia

awareness in long-standing type

1 diabetes: a multicenter 2 x 2

Ch

apte

r 1

30

factorial randomized controlled

trial comparing insulin pump

with multiple daily injections and

continuous with conventional

glucose self-monitoring

(HypoCOMPaSS). Diabetes Care 2014;

37(8): 2114-22.

66. Ly TT, Nicholas JA, Retterath A, Lim

EM, Davis EA, Jones TW. Effect of

sensor-augmented insulin pump

therapy and automated insulin

suspension vs standard insulin pump

therapy on hypoglycemia in patients

with type 1 diabetes: a randomized

clinical trial. JAMA 2013; 310(12):

1240-7.

67. Ritholz MD, Atakov-Castillo A,

Beste M, et al. Psychosocial factors

associated with use of continuous

glucose monitoring. Diabet Med

2010; 27(9): 1060-5.

68. Peters AL, Ahmann AJ, Battelino T, et

al. Diabetes Technology-Continuous

Subcutaneous Insulin Infusion

Therapy and Continuous Glucose

Monitoring in Adults: An Endocrine

Society Clinical Practice Guideline. J

Clin Endocrinol Metab 2016; 101(11):

3922-37.

69. Rubin RR, Peyrot M. Psychological

issues and treatments for people

with diabetes. J Clin Psychol 2001;

57(4): 457-78.

70. Lustman PJ, Anderson RJ, Freedland

KE, de GM, Carney RM, Clouse RE.

Depression and poor glycemic

control: a meta-analytic review of

the literature. Diabetes Care 2000;

23(7): 934-42.

71. Wild D, von MR, Brohan E,

Christensen T, Clauson P, Gonder-

Frederick L. A critical review of the

literature on fear of hypoglycemia in

diabetes: Implications for diabetes

management and patient education.

Patient Educ Couns 2007; 68(1): 10-5.

72. Barendse S, Singh H, Frier

BM, Speight J. The impact of

hypoglycaemia on quality of life and

related patient-reported outcomes in

Type 2 diabetes: a narrative review.

Diabet Med 2012; 29(3): 293-302.

73. Fisher L, Hessler D, Polonsky W,

Strycker L, Masharani U, Peters A.

Diabetes distress in adults with type

1 diabetes: Prevalence, incidence

and change over time. J Diabetes

Complications 2016; 30(6): 1123-8.

74. Tanenbaum ML, Hanes SJ, Miller

KM, Naranjo D, Bensen R, Hood KK.

Diabetes Device Use in Adults With

Type 1 Diabetes: Barriers to Uptake

and Potential Intervention Targets.

Diabetes Care 2017; 40(2): 181-7.

75. Pickup JC, Ford HM, Samsi K.

Real-time continuous glucose

monitoring in type 1 diabetes: a

qualitative framework analysis of

patient narratives. Diabetes Care

2015; 38(4): 544-50.

Journal of Diabetes Science and Technology 2016 Nov;10(6):1251-1258

Cornelis A J van Beers

J Hans DeVries

2

Continuous Glucose

Monitoring: Impact on

Hypoglycaemia

Ch

apte

r 2

34

ABSTRACT

The necessity of strict glycaemic control is unquestionable. However, hypoglycaemia

remains a major limiting factor in achieving satisfactory glucose control, and evidence

is mounting to show that hypoglycaemia is not benign. Over the past decade, evidence

has consistently shown that real-time continuous glucose monitoring improves

glycaemic control in terms of lowering glycated haemoglobin levels. However,

real-time continuous glucose monitoring has not met the expectations of the diabetes

community with regard to hypoglycaemia prevention. The earlier trials did not

demonstrate any effect on either mild or severe hypoglycaemia and the effect of

real-time continuous glucose monitoring on nocturnal hypoglycaemia was often not

reported. However, trials specifically designed to reduce hypoglycaemia in patients

with a high hypoglycaemia risk have demonstrated a reduction in hypoglycaemia,

suggesting that real-time continuous glucose monitoring can prevent hypoglycaemia

when it is specifically used for that purpose. Moreover, the newest generation of

diabetes technology currently available commercially, namely sensor-augmented

pump therapy with a (predictive) low-glucose suspend feature, has provided more

convincing evidence for hypoglycaemia prevention. This article provides an overview

of the hypoglycaemia outcomes of randomized controlled trials that investigate the

effect of real-time continuous glucose monitoring alone or sensor-augmented pump

therapy with a (predictive) low-glucose suspend feature. Furthermore, several

possible explanations are provided why trials have not shown a reduction in severe

hypoglycaemia. In addition, existing evidence is presented of real-time continuous

glucose monitoring in patients with impaired awareness of hypoglycaemia who have

the highest risk of severe hypoglycaemia.

35

INTRODUCTION

The benefits of intensive glycaemic control in reducing the microvascular and

macrovascular complications of diabetes are well established.1,2 Although strict

glycaemic control has been associated with an increased risk of hypoglycemia,1 more

recent observational data do not confirm this association.3,4 However, in daily practice,

hypoglycaemia remains the main side-effect of insulin therapy and barrier to achieving

glycaemic targets.5

The categorization of hypoglycaemic episodes is a matter of continuous debate.6

Mild hypoglycaemia is usually defined as an episode in which a person is able to

recognize and self-treat a low level of blood glucose. Severe hypoglycaemia is often

defined as a hypoglycaemic event requiring assistance of a third party.7 The American

Diabetes Association proposed a biochemical definition of hypoglycaemia as a plasma

glucose of ≤70 mg/dl (3.9 mmol/L). However, many trials use different (biochemical)

definitions of hypoglycaemia, which makes their comparison difficult. In type 1

diabetes, the mean incidence of mild hypoglycaemia is 1-2 events per patient per

week and the incidence of severe hypoglycaemia is approximately 0.1-1.5 events

per patient year. Hypoglycaemia is not benign, but has important physical and

psychosocial consequences.10,11 Hypoglycaemia interferes with many aspects of daily

life, including sleep, driving, exercise, social functioning and employment.11 In people

with type 2 diabetes with significant cardiovascular risk, hypoglycaemia probably

increases the risk of cardiovascular events, 12-14 although causality remains difficult to

prove.15 Furthermore, hypoglycaemia impairs cerebral function and might promote

permanent cognitive decline.16,17 Recurrent hypoglycaemia induces defective glucose

counterregulation and impaired awareness of hypoglycaemia (IAH).18,19 Impaired

awareness of hypoglycaemia (IAH) is associated with a three to six fold increased

risk of severe hypoglycaemia which considerably impairs their quality of life.20,21

Hypoglycaemia can also have a profound effect on psychosocial well-being and causes

fear of hypoglycemia.22-24 Healthcare costs are substantially increased because of

hypoglycemia.25 Importantly, hypoglycaemia can be fatal, with mortality estimates

ranging from 4 to 10 percent of deaths in T1DM patients diagnosed in childhood or

early adulthood and dying before the age of 40 years.26,27 A recent registry-based

observational study showed that in T1DM patients younger than 30 years, 31.4%

of deaths was caused by diabetic ketoacidosis or hypoglycemia.28 Therefore, new

treatment and monitoring strategies to prevent hypoglycaemia are a necessity.

Ch

apte

r 2

36

Although the technique of continuous glucose monitoring became available during

the late 1990s, it was not until 2006 that real-time continuous glucose monitoring (CGM)

was introduced to assist patients in their self-management.29 Present CGM systems that

are available use small minimally invasive sensors which measure interstitial glucose

levels via the glucose-oxidase reaction and translate this into blood glucose values by

means of calibrations.30,31 The CGM systems provide this information every five or ten

minutes, with a delay of approximately 5 to 15 minutes. The added value lies in the

semi-continuous display of ‘current’ glucose values, visualization of glucose trends

and the availability alarms that can be set to warn for impending hypoglycaemia or

hyperglycemia.32 First generation CGM systems were used as stand-alone devices. Next

generation CGM systems are connected to insulin pumps (sensor-augmented pump

therapy; SAPT), but do not interfere with insulin delivery automatically. These CGM

systems therefore only act as behaviour modifiers, rather than insulin dose adjustment

tools. The newest generation SAPT systems however have a (predicted) low-glucose

suspend (LGS) feature, which automatically interrupts insulin administration when

glucose falls below a pre-set threshold.33,34 This steady improvement and development

of CGM systems over the last 15 years is welcome, although to some extent it has

frustrated evaluation of the clinical evidence. In some trials the benefit of CGM itself

was studied, while other trials evaluated the combined effect of CGM and insulin

pumps (sometimes with a built-in bolus calculator or automated insulin suspension).

Continuous glucose monitoring enabled the development of new (CGM-derived)

measures to assess glycaemic control (i.e. time in target, area under the curve and

different variability measures).35,36 Most CGM trials used time below target to assess

effect of CGM on hypoglycaemia. Although time below target is a simple and easy to

understand measure, formal evidence demonstrating the usefulness of assessing time

below target compared to other measures (i.e. frequency of hypoglycaemic events), in

evaluating clinical benefit of CGM, is lacking. In this narrative review we have provided

an overview of the CGM trials and mainly focus on the hypoglycaemia outcomes.

Sensor-augmented pump therapy with (predictive) LGS will be discussed separately.

Also, we discuss trials that were performed in patients with IAH.20 Relevant articles

were identified by searching the PubMed database using the following search terms:

“continuous glucose monitoring”, “sensor-augmented pump therapy”, “low-glucose

insulin suspension”, “predictive low-glucose suspension”, “automated insulin pump

suspension”, “threshold insulin pump interruption”, “diabetes mellitus” and “type

1 diabetes”. In addition, references of selected articles were searched for additional

37

relevant articles. The closed-loop systems are beyond the scope of this review, but are

reviewed elsewhere.37

Ch

apte

r 2

38

REAL-TIME CONTINUOUS GLUCOSE MONITORING

Mild hypoglycaemia

Most randomized controlled trials (RCT) investigating the effect of CGM on glycaemia

primarily aimed at lowering HbA1c, rather than on preventing hypoglycaemia. These

trials often included patients with suboptimally controlled diabetes and evaluated

HbA1c as primary endpoint.38-48 The 2008 JDRF trial was the first landmark RCT

investigating the efficacy and safety of CGM.38 In total, 322 children, adolescents

and adults with T1DM using insulin pumps or multiple daily injections (MDI) were

randomized to receive CGM or to continue self-monitoring of blood glucose by finger

prick (SMBG) for 26 weeks. The study demonstrated a significant reduction in HbA1c

of 0,5% in adult participants. However, no significant effect was found on time spent in

hypoglycaemia. In addition, other trials comparing conventional CGM with SMBG and

focusing on HbA1c reduction either did not report on mild hypoglycaemia or did not

demonstrate any effect on mild hypoglycemia.39-41

Two RCTs compared SAPT with MDI and SMBG.42,43 In the STAR-3 trial, 485 T1DM

patients used SAPT or continued using MDI and SMBG for 1 year.42 Patients who

experienced two severe hypoglycaemic events or more in the year prior to enrolment

were excluded. HbA1c improved significantly more in the SAPT group, with a between

group difference of 0.6% (p < 0.001). However, the STAR-3 trial demonstrated no

difference in mild hypoglycaemia. These findings were supported and extended by the

EURYTHMICS trial, which evaluated 83 patients for six months and found an impressive

HbA1c reduction in the SAPT group (-1,2%, p < 0.001), but again no significant reduction

was observed in time spent in hypoglycaemia or the number of mild hypoglycaemic

events.43

Several RCTs investigated the incremental effect of CGM when using an insulin

pump.44-48 Overall, these studies either did not find a significant44,45 or relevant46

reduction in mild hypoglycaemia or did not report on the occurrence of mild

hypoglycaemic episodes.47 However, the SWITCH Study Group did find a significant

effect of adding CGM to insulin pump therapy on time spent in hypoglycemia.48

This cross-over trial randomized 153 children and adults with T1DM using CSII to a

Sensor On or Sensor Off arm for 6 months. After a washout of 4 months, participants

switched to the other arm. During CGM use, less time was spent in hypoglycaemia,

with 19 min/day <70 mg/dl (3.9 mmol/L) in the Sensor On arm and 31 min/day <70 mg/

dl (3.9 mmol/L) in the Sensor Off arm (p = 0.009). In addition, the average daily AUC

<70 mg/dl (3.9 mmol/L) was significantly lower in the Sensor On arm group. Notably,

39

this cross-over trial gathered 8 weeks of blinded CGM values. Other trials often used

less than 14 days of blinded CGM data to analyse CGM-derived outcomes, such as time

spent in hypoglycaemia or mild hypoglycaemia event rate.38,42,43 It is possible that this

relatively small amount of blinded CGM data lacked power to demonstrate between

group differences.

Few RCTs evaluating the efficacy of CGM primarily aimed at hypoglycaemia

prevention.49,50 Interestingly, these studies did demonstrate a significant reduction in

mild hypoglycaemia. The 2009 JDRF trial examined the effect of CGM versus SMBG in

129 adults and children with T1DM and a HbA1c <7.0%.49 Time spent in hypoglycaemia

decreased significantly in the CGM group from 91 min/day ≤70 mg/dl (3.9 mmol/L) at

baseline to 54 min/day ≤70 mg/dl (3.9 mmol/L) at 26 weeks (p = 0.002). Marginally nonsig-

nificant, the mild hypoglycaemic event rate was less pronounced in the CGM group,

with 0.25 events/day versus 0.47 events/day in the control group (P = 0.07). Moreover,

in 2011, Battelino et al. assessed the impact of RT-CGM versus SMBG specifically on

hypoglycaemia in 120 children and adults with T1DM and a HbA1c <7.5%.50 The

authors reported less time spent in hypoglycaemia in the CGM group compared with

the control group (0.91 hours/day <70 mg/dl (3.9 mmol/L) vs. 1.6 hours/day <70 mg/dl

(3.9 mmol/L), respectively; p = 0.01). Furthermore, the number of mild hypoglycaemic

events per day was lower in the CGM group (0.53 events/day in the CGM group vs. 0.76

events/day in the control group, p = 0.08).

Nocturnal hypoglycaemia

Nocturnal hypoglycaemia is of major concern to people with T1DM. Studies using

CGM report a prevalence of nocturnal hypoglycaemia of up to 68%.51-53 In the DCCT, half

of the severe hypoglycaemic events occurred during sleep.54 In addition, in children,

up to 75% of hypoglycaemic events associated with seizures or coma occur at night

when counterregulatory responses are impaired.54-56 Furthermore, the “dead-in-bed”

syndrome accounts for approximately 6% of all deaths in people with T1DM under the

age of 40 years, which is probably related to severe nocturnal hypoglycemia.57

Continuous glucose monitoring studies reporting on nocturnal hypoglycaemia

should be interpreted with caution due to concerns around the accuracy of glucose

sensors at night (i.e. due to compression artefacts, disconnections and lack of

calibrations at night).58-60

The impact of CGM on nocturnal hypoglycaemia has seldom been reported.29,50 In a

study by Garg et al., nocturnal hypoglycaemia (<55 mg/dl (3.1 mmol/L)) was reduced by

Ch

apte

r 2

40

38% in the display on group compared with the control group (p < 0.001).29 In addition,

the trial of Battelino et al. reported significantly lower hypoglycaemic excursions during

the night in the CGM group compared with control (0.13 vs. 0.19 excursions/night <55

mg/dl (3.1 mmol/L), p = 0.01 and 0.21 vs. 0.30 excursions/night <63 mg/dl (3.5 mmol/L),

p = 0.009).50 Other trials investigating the efficacy of CGM either did not evaluate the

effect on nocturnal hypoglycaemia, or did not report it. Future studies evaluating the

effect of CGM on hypoglycaemia should report nocturnal hypoglycaemia.

Severe hypoglycaemia

Continuous glucose monitoring was expected to reduce severe hypoglycemia.61

Unfortunately, evidence supporting this belief is still lacking. No RCTs investigating

CGM showed a significant decrease in severe hypoglycaemia (Table 1). One of the

earliest trials even reported a significant increase of severe hypoglycaemia in the CGM

group.46

Some meta-analyses are performed comparing severe hypoglycaemic event rates

during CGM versus SMBG.62,63 In 2011, Pickup et al. performed an individual patient

level meta-analysis.62 The overall severe hypoglycaemia incidence rate ratio on SMBG

compared with CGM was 1.40 (0.87 – 2.25, p = 0.17). These findings were supported

by the Cochrane Collaboration in 2012, which also found no difference in incidence

rates of severe hypoglycaemia between CGM and SMBG (risk ratio 1.05 [95% CI 0.63 –

1.77]).63

Several explanations have been put forward to explain why CGM does not seem

to prevent severe hypoglycaemia, or, why trials are unable to demonstrate this.

Importantly, none of the trials had sufficient power to demonstrate a difference in

severe hypoglycaemia. Moreover, most trials were designed to lower HbA1c instead

of preventing (severe) hypoglycaemia and in some trials, patients with recent severe

hypoglycaemia or IAH were excluded.42,45,48 In these trials, patients and study staff

may have been less focused on preventing hypoglycaemia. Since CGM devices act only

as behaviour modifiers, the focus of patient and caregiver to reduce hypoglycaemia

is of major importance to perceive this goal. Furthermore, although the accuracy of

CGM systems have steadily improved over the last decade64,65, the performance of CGM

devices is still poorest in the hypoglycaemic range, which may hinder its ability to

provide an adequate alarm to prevent severe hypoglycaemia. Also, qualitative studies

show that frequent (inadequate) alarms irritate the user and are a major barrier to

the effective use of CGM.66,67 Hypoglycaemia-induced cognitive decline and sleep may

cause inadequate responses to alarms.68,69

41

IMPAIRED AWARENESS OF HYPOGLYCAEMIA

Whether CGM can prevent hypoglycaemia in patients with IAH, either directly or by

improving hypoglycaemia awareness, has yet to be established. In 2011, a hyperinsu-

linemic hypoglycaemic clamp study by Ly et al. showed that 4 weeks of CGM improved

epinephrine responses in young T1DM patients with IAH, suggesting that IAH can be

restored in adolescents by using CGM.70 This finding was not supported by a larger

trial performed by the same study group.71 In 2014, Little et al. evaluated in the

HypoCOMPaSS trial whether hypoglycaemia awareness can be improved and severe

hypoglycaemia can be prevented by using strategies available in routine practice,

including CGM.72 This randomized controlled trial had a 24-week 2 × 2 factorial design,

comparing CSII with MDI and CGM with SMBG. All participants received written insulin

titration guidelines, educational sessions, weekly telephone consultations and monthly

visits in order to achieve rigorous avoidance of biochemical hypoglycaemia. After 24

weeks, hypoglycaemia awareness scores measured according to the method of Gold et

al.21 (scale of 1 to 7) had improved from 5.1 to 4.1 ( P = 0.0001), without between-group

differences. The clinical relevance of this improvement in hypoglycaemia awareness

is unknown. Although the improvement in hypoglycaemia awareness scores was

accompanied by a significant reduction in severe hypoglycaemia, from 8.9 events per

patient-year at baseline to 0.8 events per patient-year after 24 weeks, this reduction

in severe hypoglycaemia was probably caused by the insulin adjustment algorithm,

education, frequent telephone consultations and consultations rather than the

improvement in hypoglycaemia awareness. The authors did not demonstrate any

difference in severe hypoglycaemia or time spent in hypoglycaemia between CGM and

SMBG, which is also most likely explained by a floor effect, with maximal reduction

already attained by this intensive guidance. Whether such intensive guidance is

feasible in routine clinical practice is under debate.73,74

The first observational study performed in patients with IAH demonstrated a

clear reduction of SH with CGM use, without change in hypoglycaemia awareness

scores75, addressing the need for further interventional studies in patients with IAH.

A randomized control trial (RCT) investigating the effects of CGM in these patients is

currently being conducted.76

Ch

apte

r 2

42

Stu

dy/

firs

t a

uth

or

Pa

rtic

ipa

nts

(n

)B

ase

lin

e H

bA

1cD

ura

tio

nC

om

pa

riso

n(i

nte

rven

tio

n v

s. c

on

tro

l)

Ou

tco

me

Hb

A1c

Mil

d

hyp

ogl

yca

emia

No

ctu

rna

l h

ypo

glyc

aem

iaS

ever

e h

ypo

glyc

ae-

mia

(n

)

Co

nti

nu

ou

s G

luco

se M

on

ito

rin

g

Focu

s: g

lyca

ted

hae

mog

lob

in r

edu

ctio

n

JDR

F 2

00838

322

7.5-

10%

6 m

onth

sC

GM

vs.

SM

BG

↓=

?14

vs.

11

Gu

ard

Con

trol

3916

2≥8

.1%

3 m

onth

sC

GM

vs.

SM

BG

↓?

?2

vs. 0

Riv

elin

e et

al.

4025

7≥8

.0%

12 m

onth

sC

GM

vs.

SM

BG

↓?

?37

vs.

15

Dir

ecN

et S

tud

y G

rou

p41

146

≥7.0

%6

mon

ths

CG

M v

s. S

MB

G=

=?

3 vs

. 6

STA

R-3

4248

57.

4-9.

5%12

mon

ths

SAP

T v

s. M

DI

+ SM

BG

↓=

?32

vs.

27

EU

RY

TH

MIC

S4383

≥8.2

%6

mon

ths

SAP

T v

s. M

DI

+ SM

BG

↓=

?4

vs. 1

Rea

lTre

nd

4413

2≥8

.0%

6 m

onth

sSA

PT

vs.

CSI

I +

SMB

G=

=?

1 vs

. 0

ASA

P45

62≤8

.5%

3 m

onth

sSA

PT

vs.

CSI

I +

SMB

G↓

=?

0 vs

. 0

STA

R-1

4614

6≥7

.5%

6 m

onth

sSA

PT

vs.

CSI

I +

SMB

G=

=?

11 v

s. 3

*

ON

SET

4716

011

.3%

(m

ean

)12

mon

ths

SAP

T v

s. C

SII

+ SM

BG

=?

?0

vs. 4

SWIT

CH

4815

37.

5-9.

5%6

mon

ths

SAP

T v

s. C

SII

+ SM

BG

↓↓

?4

vs. 2

Focu

s: h

ypog

lyca

emia

red

uct

ion

JDR

F 2

00949

129

<7,0

%6

mon

ths

CG

M v

s. S

MB

G↓

↓?

9 vs

. 10

Bat

teli

no

et a

l. 2

01150

120

<7.5

%6

mon

ths

CG

M v

s. S

MB

G↓

↓↓

0 vs

. 0

Hyp

oCO

MP

aSS72

968.

2% (

mea

n)

6 m

onth

sC

GM

vs.

SM

BG

==

?46

vs.

44

43

Sen

sor-

au

gmen

ted

pu

mp

th

era

py

wit

h (

pre

dic

ted

) lo

w-g

luco

se s

usp

ensi

on

ASP

IRE

In

-Hom

e3324

75.

8-10

.0%

3 m

onth

sSA

PT

+ L

GS

vs. S

AP

T=

↓↓

0 vs

. 4

Ly

et a

l.71

95

≤8.5

%6

mon

ths

SAP

T +

LG

S vs

. CSI

I +

SMB

G=

↓↓

35 v

s. 1

3‡

Maa

hs

et a

l.34

45≤8

.0%

42 n

igh

tsSA

PT

+ P

LG

S vs

. SA

PT

?

↓↓

0 vs

. 0

Bu

ckin

gham

et

al.80

81

≤8.5

%42

nig

hts

SAP

T +

PL

GS

vs. S

AP

T

?↓

↓0

vs. 0

↓Sig

nif

ican

t re

du

ctio

n i

n o

utc

ome

mea

sure

in

in

terv

enti

on g

rou

p v

s. c

ontr

ol g

rou

p. =

Non

-sig

nif

ican

t b

etw

een

-gro

up

dif

fere

nce

. ? R

esu

lts

not

rep

orte

d. *

P <

0.0

5. ‡

Inci

den

ce r

ate

per

100

pa-

tien

t-m

onth

s ad

just

ed f

or b

asel

ine:

9.5

vs.

34.

2, p

< 0

.001

. CG

M, c

onti

nu

ous

glu

cose

mon

itor

ing.

CSI

I, c

onti

nu

ous

sub

cuta

neo

us

insu

lin

inje

ctio

n. L

GS,

low

-glu

cose

su

spen

sion

. PL

GS,

pre

dic

tive

low

-glu

cose

su

spen

sion

MD

I, m

ult

iple

dai

ly in

ject

ion

s. S

AP

T, s

enso

r-au

gmen

ted

pu

mp

th

erap

y. S

MB

G, s

elf-

mon

itor

ing

of b

lood

glu

cose

.

Ta

ble

1. O

ver

vie

w o

f C

GM

ra

nd

om

ised

co

ntr

oll

ed t

ria

ls i

n t

ype

1 d

iab

etes

Ch

apte

r 2

44

SENSOR-AUGMENTED PUMP THERAPY WITH (PREDIC-TIVE) LOW-GLUCOSE SUSPENSION

Mild and nocturnal hypoglycaemia

Several feasibility studies evaluating the LGS feature have demonstrated less

time spent in hypoglycaemia and fewer hypoglycaemic episodes77, less nocturnal

hypoglycaemia in those at greatest risk78 and shorter duration of hypoglycaemic

episodes79 without an increased risk of ketoacidosis or hyperglycaemia.

In 2013, the ASPIRE In-Home Study Group evaluated the effect of SAPT with LGS,

compared with SAPT alone, on nocturnal hypoglycaemia and glycated haemoglobin

levels.33 Patients were eligible if they experienced ≥2 nocturnal hypoglycaemic

events during the 2-week run-in phase. After 3 months the mean AUC for nocturnal

hypoglycaemic events was 980 mg/dl × minutes (54.4 mmol/L × minutes) in the LGS

group and 1568 mg/dl × minutes (87 mmol/L × minutes) in the control group, a 38%

reduction (p < 0.001). In addition, the frequency of nocturnal hypoglycaemic events

was significantly reduced by 31.8% in the LGS group (p < 0.001). The time spent in

hypoglycaemia was also significantly lower in the LGS group. There were no severe

hypoglycaemic events in the LGS group and 4 severe hypoglycaemic events in the

control group.

The In Home Closed Loop Study Group assessed the safety and effectiveness of a

predictive low-glucose suspend feature in a 42-night in-home randomized trial.34

Each night, the 45 T1DM patients were randomly assigned to having the predictive

low-glucose suspend feature on (intervention) or off (SAPT only, control). The

proportion of nights in which ≥1 sensor value ≤60 mg/dl (3.3 mmol/L) occurred was

analysed as primary outcome. At least 1 sensor value ≤60 mg/dl (3.3 mmol/L) occurred

during 21% of the intervention nights, compared with 33% of the control nights (p

< 0.001). In addition, the intervention reduced the duration, frequency and AUC of

nocturnal hypoglycaemia significantly. These findings were accompanied by the

results of a similar RCT of the In Home Closed Loop Study Group performed in children

with T1DM.80 In both trials, morning ketosis did not differ between the intervention

and control nights. Mean overnight and morning glucose values were slightly higher

during and after the intervention nights.

These data suggest that using (predictive) LGS in addition to SAPT is safe and effective

in reducing the size (AUC per hypoglycaemic event) and the frequency of (nocturnal)

hypoglycaemic events.

45

Severe hypoglycaemia

In 2013, Ly et al. investigated the effect of SAPT with LGS on the combined frequency

of moderate (defined as a hypoglycaemic event requiring third party assistance) and

severe (defined as a hypoglycaemic event resulting in seizure or coma) hypoglycaemia

in T1DM patients with IAH.71 At baseline, the combined moderate and severe

hypoglycaemia rate was significantly higher in the LGS group, with 129.6 events

per 100 patient-months, compared with 20.7 events per 100 patient-months in the

control group. After 6 months, the incidence rate of combined moderate and severe

hypoglycaemia, adjusted for baseline rates, was significantly lower in the LGS group

compared with the control group (9.5 events per 100 patient-months vs. 34.2 events per

100 patient-months, respectively), resulting in an incidence rate ratio of 3.6 in favour of

the LGS group (P < 0.001). In addition, combined daytime and night-time time spent in

hypoglycaemia was significantly lower in the LGS group. However, when two outliers

with the highest baseline event rates of moderate hypoglycaemia were excluded, the

primary outcome lost significance (rate ratio 2.2, P = 0.08). Furthermore, the German

Institute for Quality and Efficiency in Health Care (IQWiG) reanalysed the data from the

Ly study, questioned its methodological quality and could not confirm its conclusions.81

They argued that hypoglycaemia rates are problematic as an endpoint, mainly because

statistical tests require independent observations and therefore the patient should be

the statistical unit, not an event, as hypoglycaemia is known to cluster in a subgroup

of patients.4,82 Indeed, the number of patients affected by severe hypoglycaemia was 3

out of 45 in the control group versus 0 out of 41 in the intervention group and analysed

with the patient as a statistical unit this difference is not significant.81 Nevertheless,

event rates in the way as analysed by Ly et al. are an accepted endpoint in diabetes

trials. In addition, by excluding the two outliers, Ly et al. explored the effect of extreme

values on the results and the issue of clustering. Furthermore, it should be pointed

out that the authors pre-specified their statistical analyses and the journal’s statistical

advisor agreed to their appropriateness. Finally, from a clinical point of view, severe

hypoglycaemia is a clinically relevant outcome. We therefore find the results of this trial

promising, since it is the first trial to demonstrate a reduction in severe hypoglycaemia

by CGM. Future trials attempting to confirm the findings of the Ly study could consider

including a sufficient number of patients experiencing severe hypoglycaemia, in order

to show a reduction in the proportion of patients affected, instead of demonstrating

a reduction of the incidence of severe hypoglycaemia only in those with the highest

incidence rates.

Ch

apte

r 2

46

CONCLUSION

Continuous glucose monitoring/SAPT can decrease HbA1c without increasing

hypoglycaemia. Most trials investigating the effect of CGM in T1DM patients however,

do not demonstrate a reduction in (severe) hypoglycaemia. But, during these trials

patients and investigators were possibly more focused on improving glycaemic

control by lowering glycated haemoglobin values than on preventing hypoglycaemia.

Since continuous glucose monitoring devices without low glucose suspension

only act as behaviour modifiers, they can only prevent hypoglycaemia through

CGM-induced behavioural changes. In order to incorporate these behavioural changes

into the self-management of patients, patients and caregivers must be focused on

preventing hypoglycaemia. This is strengthened by the fact that the trials focusing on

hypoglycaemia prevention also show a reduction in mild hypoglycaemia, suggesting

that CGM is able to prevent hypoglycaemia when it is used for that purpose. In our

opinion, this behaviour factor is often undervalued. Trials investigating the effect of

SAPT with LGS more convincingly show its benefit in mild, nocturnal and possibly

even severe hypoglycaemia prevention. Although self-management will always be an

important factor in the management of T1DM, this underscores the need for further

replacement of self-management by automated insulin therapy.

47








329(14): 977-86.





J Med 2005; 353(25): 2643-53.






971-8.






Metab Res Rev 2004; 20(6): 479-86.





937-48.

6. Frier BM. Defining hypoglycaemia:

what level has clinical relevance?

Diabetologia 2009; 52(1): 31-4.

7. American Diabetes Asoociation






Diabetes Care 2005; 28(5): 1245-9.

8. Ostenson CG, Geelhoed-Duijvestijn P,

Lahtela J, Weitgasser R, Markert JM,

Pedersen-Bjergaard U. Self-reported

non-severe hypoglycaemic events

in Europe. Diabet Med 2014; 31(1):

92-101.

9. MacLeod KM, Hepburn DA, Frier BM.

Frequency and morbidity of severe

hypoglycaemia in insulin-treated

diabetic patients. Diabet Med 1993;

10(3): 238-45.




2014; 10(12): 711-22.




2008; 24(2): 87-92.

12. Duckworth W, Abraira C, Moritz T,

et al. Glucose control and vascular

complications in veterans with type

2 diabetes. N Engl J Med 2009; 360(2):

129-39.

13. Gerstein HC, Miller ME, Byington

RP, et al. Effects of intensive glucose

lowering in type 2 diabetes. N Engl J

Med 2008; 358(24): 2545-59.

14. Patel A, MacMahon S, Chalmers J, et

al. Intensive blood glucose control

and vascular outcomes in patients

with type 2 diabetes. N Engl J Med

2008; 358(24): 2560-72.

15. Bonds DE, Miller ME, Bergenstal

REFERENCES

Ch

apte

r 2

48

RM, et al. The association between

symptomatic, severe hypoglycaemia

and mortality in type 2 diabetes:

retrospective epidemiological

analysis of the ACCORD study. BMJ

2010; 340: b4909.




7(5): 493-503.

17. Feinkohl I, Aung PP, Keller M, et al.

Severe hypoglycemia and cognitive

decline in older people with type

2 diabetes: the Edinburgh type 2

diabetes study. Diabetes Care 2014;

37(2): 507-15.

18. Davis SN, Shavers C, Mosque-

da-Garcia R, Costa F. Effects of

differing antecedent hypoglycemia

on subsequent counterregulation

in normal humans. Diabetes 1997;

46(8): 1328-35.

19. Cryer PE. Diverse causes of hypogly-

cemia-associated autonomic failure

in diabetes. N Engl J Med 2004;

350(22): 2272-9.





Med 2008; 25(4): 501-4.






17(7): 697-703.

22. Gonder-Frederick L, Nyer M,

Shepard JA, Vajda K, Clarke W.

Assessing fear of hypoglycemia in

children with Type 1 diabetes and

their parents. Diabetes Manag (Lond)

2011; 1(6): 627-39.

23. Gonder-Frederick LA, Clarke WL,

Cox DJ. The Emotional, Social, and

Behavioral Implications of Insulin-

Induced Hypoglycemia. Semin Clin

Neuropsychiatry 1997; 2(1): 57-65.

24. Gonder-Frederick LA, Cox DJ, Bobbitt

SA, Pennebaker JW. Mood changes

associated with blood glucose

fluctuations in insulin-dependent

diabetes mellitus. Health Psychol

1989; 8(1): 45-59.

25. Foos V, Varol N, Curtis BH, et al.

Economic impact of severe and

non-severe hypoglycemia in patients

with Type 1 and Type 2 diabetes in

the United States. J Med Econ 2015;

18(6): 420-32.








young adults. Diabetes Care 2008;

31(5): 922-6.





since 1989. Diabetologia 2007; 50(12):

2439-42.




Med 2014; 371(21): 1972-82.


49





Care 2006; 29(1): 44-50.






Technol 2009; 3(5): 1207-14.







769-79.







105-13.





369(3): 224-32.

34. Maahs DM, Calhoun P, Buckingham

BA, et al. A randomized trial of a

home system to reduce nocturnal

hypoglycemia in type 1 diabetes.

Diabetes Care 2014; 37(7): 1885-91.

35. DeVries JH. Glucose variability:

where it is important and how to

measure it. Diabetes 2013; 62(5):

1405-8.

36. Siegelaar SE, Holleman F, Hoekstra

JB, DeVries JH. Glucose variability;

does it matter? Endocr Rev 2010;

31(2): 171-82.

37. Kropff J, DeVries JH. Continuous

Glucose Monitoring, Future Products,

and Update on Worldwide Artificial

Pancreas Projects. Diabetes Technol

Ther 2016; 18 Suppl 2: S253-S63.

38. Tamborlane WV, Beck RW, Bode BW,

et al. Continuous glucose monitoring

and intensive treatment of type 1

diabetes. N Engl J Med 2008; 359(14):

1464-76.

39. Deiss D, Bolinder J, Riveline JP, et al.

Improved glycemic control in poorly

controlled patients with type 1

diabetes using real-time continuous

glucose monitoring. Diabetes Care

2006; 29(12): 2730-2.

40. Riveline JP, Schaepelynck P, Chaillous

L, et al. Assessment of patient-led

or physician-driven continuous

glucose monitoring in patients with

poorly controlled type 1 diabetes

using basal-bolus insulin regimens:

a 1-year multicenter study. Diabetes

Care 2012; 35(5): 965-71.

41. Mauras N, Beck R, Xing D, et al. A

randomized clinical trial to assess

the efficacy and safety of real-time

continuous glucose monitoring in the

management of type 1 diabetes in

young children aged 4 to <10 years.

Diabetes Care 2012; 35(2): 204-10.

42. Bergenstal RM, Tamborlane WV,

Ahmann A, et al. Effectiveness of

sensor-augmented insulin-pump

therapy in type 1 diabetes. N Engl J

Med 2010; 363(4): 311-20.


Ch

apte

r 2

50

Bruttomesso D, et al. Sensor-aug-

mented pump therapy lowers




28(10): 1158-67.

44. Raccah D, Sulmont V, Reznik Y, et

al. Incremental value of continuous

glucose monitoring when starting

pump therapy in patients with

poorly controlled type 1 diabetes: the

RealTrend study. Diabetes Care 2009;

32(12): 2245-50.

45. O'Connell MA, Donath S, O'Neal DN,

et al. Glycaemic impact of patient-led

use of sensor-guided pump therapy

in type 1 diabetes: a randomised

controlled trial. Diabetologia 2009;

52(7): 1250-7.

46. Hirsch IB, Abelseth J, Bode BW,

et al. Sensor-augmented insulin

pump therapy: results of the first

randomized treat-to-target study.


377-83.

47. Kordonouri O, Pankowska E, Rami

B, et al. Sensor-augmented pump

therapy from the diagnosis of

childhood type 1 diabetes: results of

the Paediatric Onset Study (ONSET)

after 12 months of treatment.

Diabetologia 2010; 53(12): 2487-95.

48. Battelino T, Conget I, Olsen B, et al.

The use and efficacy of continuous

glucose monitoring in type 1 diabetes

treated with insulin pump therapy:

a randomised controlled trial.

Diabetologia 2012; 55(12): 3155-62.

49. Beck RW, Hirsch IB, Laffel L, et al.

The effect of continuous glucose

monitoring in well-controlled type 1


1378-83.

50. Battelino T, Phillip M, Bratina N,

Nimri R, Oskarsson P, Bolinder

J. Effect of continuous glucose

monitoring on hypoglycemia in type

1 diabetes. Diabetes Care 2011; 34(4):

795-800.

51. Ahmet A, Dagenais S, Barrowman NJ,

Collins CJ, Lawson ML. Prevalence of

nocturnal hypoglycemia in pediatric

type 1 diabetes: a pilot study using

continuous glucose monitoring. J

Pediatr 2011; 159(2): 297-302.

52. Guillod L, Comte-Perret S, Monbaron

D, Gaillard RC, Ruiz J. Nocturnal

hypoglycaemias in type 1 diabetic

patients: what can we learn with

continuous glucose monitoring?

Diabetes Metab 2007; 33(5): 360-5.

53. Wentholt IM, Maran A, Masurel N,

Heine RJ, Hoekstra JB, DeVries JH.

Nocturnal hypoglycaemia in Type

1 diabetic patients, assessed with

continuous glucose monitoring:

frequency, duration and associations.

Diabet Med 2007; 24(5): 527-32.

54. The DCCT Research Group.

Epidemiology of severe

hypoglycemia in the diabetes control

and complications trial. Am J Med

1991; 90(4): 450-9.

55. Matyka KA, Crowne EC, Havel

PJ, Macdonald IA, Matthews D,

Dunger DB. Counterregulation

during spontaneous nocturnal

hypoglycemia in prepubertal

51

children with type 1 diabetes.

Diabetes Care 1999; 22(7): 1144-50.

56. Davis EA, Keating B, Byrne GC,

Russell M, Jones TW. Hypoglycemia:

incidence and clinical predictors in

a large population-based sample of

children and adolescents with IDDM.

Diabetes Care 1997; 20(1): 22-5.

57. Sovik O, Thordarson H. Dead-in-bed

syndrome in young diabetic patients.


B40-B2.

58. Facchinetti A, Del FS, Sparacino G,

Cobelli C. Modeling Transient Discon-

nections and Compression Artifacts

of Continuous Glucose Sensors.


264-72.

59. Bay C, Kristensen PL, Pedersen-

Bjergaard U, Tarnow L,

Thorsteinsson B. Nocturnal

continuous glucose monitoring:

accuracy and reliability of

hypoglycemia detection in patients

with type 1 diabetes at high risk

of severe hypoglycemia. Diabetes

Technol Ther 2013; 15(5): 371-7.

60. Calhoun P, Lum J, Beck RW, Kollman

C. Performance comparison of the

medtronic sof-sensor and enlite

glucose sensors in inpatient studies

of individuals with type 1 diabetes.


758-61.

61. Wentholt IM, Hoekstra JB, DeVries

JH. Continuous glucose monitors: the

long-awaited watch dogs? Diabetes

Technol Ther 2007; 9(5): 399-409.














CD008101.

64. Kovatchev B, Anderson S,

Heinemann L, Clarke W. Comparison

of the numerical and clinical

accuracy of four continuous glucose

monitors. Diabetes Care 2008; 31(6):

1160-4.

65. Kropff J, Bruttomesso D, Doll W, et al.

Accuracy of two continuous glucose

monitoring systems: a head-to-head

comparison under clinical research

centre and daily life conditions.

Diabetes Obes Metab 2015; 17(4):

343-9.






2015; 38(4): 544-50.





2010; 27(9): 1060-5.

68. Buckingham B, Block J, Burdick J,

et al. Response to nocturnal alarms

Ch

apte

r 2

52

using a real-time glucose sensor.


440-7.

69. Cryer PE, Davis SN, Shamoon H.

Hypoglycemia in diabetes. Diabetes

Care 2003; 26(6): 1902-12.

70. Ly TT, Hewitt J, Davey RJ, Lim EM,

Davis EA, Jones TW. Improving

epinephrine responses in

hypoglycemia unawareness with

real-time continuous glucose

monitoring in adolescents with type


50-2.









1240-7.











37(8): 2114-22.

73. Edelman SV. Comment on Little

et al. Recovery of hypoglycemia








(HypoCOMPaSS). Diabetes Care

2014;37:2114-2122. Diabetes Care

2014; 37(12): e270-e1.


E, et al. Response to comment on

Little et al. Recovery of hypoglycemia








(HypoCOMPaSS). Diabetes Care

2014;37:2114-2122. Diabetes Care

2014; 37(12): e272-e3.

75. Choudhary P, Ramasamy S, Green L,

et al. Real-time continuous glucose

monitoring significantly reduces

severe hypoglycemia in hypoglyce-

mia-unaware patients with type 1


4160-2.

76. van Beers CA, Kleijer SJ, Serne

EH, et al. Design and rationale of

the IN CONTROL trial: the effects

of real-time continuous glucose

monitoring on glycemia and

quality of life in patients with type

1 diabetes mellitus and impaired

awareness of hypoglycemia. BMC

Endocr Disord 2015; 15: 42.

77. Danne T, Kordonouri O, Holder M,

et al. Prevention of hypoglycemia by

using low glucose suspend function

53

in sensor-augmented pump therapy.


1129-34.

78. Choudhary P, Shin J, Wang Y, et

al. Insulin pump therapy with

automated insulin suspension in

response to hypoglycemia: reduction

in nocturnal hypoglycemia in those

at greatest risk. Diabetes Care 2011;

34(9): 2023-5.

79. Garg S, Brazg RL, Bailey TS, et

al. Reduction in duration of

hypoglycemia by automatic

suspension of insulin delivery: the

in-clinic ASPIRE study. Diabetes

Technol Ther 2012; 14(3): 205-9.

80. Buckingham BA, Raghinaru

D, Cameron F, et al. Predictive

Low-Glucose Insulin Suspension

Reduces Duration of Nocturnal

Hypoglycemia in Children Without

Increasing Ketosis. Diabetes Care

2015; 38(7): 1197-204.

81. Heinemann L, Hermanns N. IQWiG

Reanalyzes and Raises Questions

About an Article by Ly et al Which

Concluded Low Glucose Suspend

Is Very Beneficial. J Diabetes Sci

Technol 2015.

82. Windeler J, Lange S. Events per

person year--a dubious concept. BMJ

1995; 310(6977): 454-6.

BMC Endocrine Disorders 2015 Aug 21;15:42

Cornelis A J van Beers, Susanne J Kleijer, Erik H Serné,

Petronella H L MGeelhoed-Duijvestijn, Frank J Snoek,

Mark H H Kramer, Michaela Diamant

3

Design and rationale of the

IN CONTROL trial: the effects


monitoring on glycaemia and




Ch

apte

r 3

56

ABSTRACT

Background

Hypoglycaemia is the main side effect of intensified insulin therapy in type 1 diabetes

and recognized as a limitation in achieving glycaemic targets. Patients with impaired

awareness of hypoglycaemia have a threefold to sixfold increased risk of severe

hypoglycaemia. Real-time continuous glucose monitoring may help patients with

type 1 diabetes to achieve better glycaemic control with less hypoglycaemic episodes.

Accordingly, one may hypothesize that particularly type 1 diabetes mellitus patients

with impaired awareness of hypoglycaemia will profit most from this technology with

improvements in their quality of life. However, this has not yet been established. This

trial aims to study the effect of real-time continuous glucose monitoring on glycaemia

and quality of life specifically in type 1 diabetes mellitus patients with established

impaired awareness of hypoglycaemia.

Methods/design

This is a two-centre, randomised, crossover trial with a 12-week wash-out period

in between intervention periods. A total of 52 type 1 diabetes mellitus patients

with impaired awareness of hypoglycaemia according to Gold et al. criteria will be

randomised to receive real-time continuous glucose monitoring or blinded continuous

glucose monitoring for 16 weeks. After a wash-out period, patients will cross over

to the other intervention. The primary outcome measure is time spent in normogly-

caemia. Secondary outcomes include (diabetes-specific) markers of quality of life and

other glycaemic variables.

Discussion

It remains unclear whether patients with type 1 diabetes and impaired awareness of

hypoglycaemia benefit from real-time continuous glucose monitoring in real-life. This

study will provide insight into the potential benefits of real-time continuous glucose

monitoring in this patient population.

57

BACKGROUND

Type 1 diabetes mellitus (T1DM) constitutes about 10-15% of total diabetes rates and

its incidence has been increasing worldwide at an alarming rate of 3-5% per year.1 T1DM

is associated with an increased risk of microvascular and macrovascular co-morbidities.

The Diabetes Control and Complication Trial/Epidemiology of Diabetes Interventions

and Complications (DCCT/EDIC) studies have shown that intensive versus conventional

glycaemic control results in a reduction of microvascular2 but also macrovascular

complications.3 However, the DCCT also demonstrated that intensive treatment

associates with a considerably increased rate of hypoglycaemia.2 With incidences of

approximately 2 per patient per week4-6 for mild (i.e. self-treated) hypoglycaemia,

and 0.1-1.5 per patient year2,5-9 for severe hypoglycaemia, hypoglycaemia is both the

main limitation in achieving glycaemic targets and the main side effect of intensified

insulin therapy in T1DM.10,11 The American Diabetes Association (ADA) defines severe

hypoglycaemia as an event requiring assistance of a third party, with symptoms of

neuroglycopenia and recovery of neurological symptoms after restoration of plasma

glucose.12

Hypoglycaemia is both a physical and psychological burden.13,14 It can interfere

with every aspect of daily life, such as sleep, exercise, driving or otherwise travelling,

but also social interactions and even employment.13 Patients worry about having

episodes of (severe) hypoglycaemia or getting the late complications of T1DM,6

and there is always the possibility of a hypoglycaemic episode leading to a coma.6

Recurrent hypoglycaemia may promote the development of impaired awareness of

hypoglycaemia (IAH) by decreasing the glycaemic threshold in the brain required for

the activation of the autonomic system.15 Consequently, by the time the glucose level is

low enough to elicit symptoms, patients fail to recognize them due to neuroglycopenia.

To date, no consensus on a satisfactory definition of IAH exists. We defined IAH as

the diminished ability to recognize the onset of hypoglycaemia.16 Hypoglycaemia

awareness can be assessed by using self-rating questionnaires. The Gold method

consists of one question: “do you know when hypoglycaemia is commencing?”. The

state of hypoglycaemia awareness is then assessed by using a 7-point visual analogue

scale, with 1 representing “always aware” and 7 representing “never aware”. A score

of ≥4 suggests impaired awareness of hypoglycaemia.17 The Clarke method consists

of eight questions identifying the respondents exposure to episodes of (mild and

severe) hypoglycaemia and examining the glycaemic threshold for hypoglycaemia.

Ch

apte

r 3

58

Impaired awareness of hypoglycaemia is implied if the respondent has a score of

≥4.18 Both scoring systems show good performance in adults.19 Impaired awareness

of hypoglycaemia renders patients at a threefold to sixfold increased risk of SH which

considerably hampers their quality of life.13,17,20 In addition, hypoglycaemia can be

fatal, with hypoglycaemia mortality estimates ranging from 4 to 10 percent of deaths

of patients with type 1 diabetes.21,22 Furthermore, a recent observational study showed

that T1DM patients with an glycated haemoglobin level of 6.9% or lower had a twice

increased risk of death from any cause and cardiovascular causes compared with

matched controls.23

In 2006, real-time continuous glucose monitoring (RT-CGM) was introduced to assist

patients in their self-management of blood glucose.24 Real-time continuous glucose

monitoring systems measure interstitial glucose levels and provide this information

every five or ten minutes, with a delay of approximately 8 to 15 minutes.25-27 The added

value lies in the display of trends and alarms that can be set to warn for impending

hypo- or hyperglycaemia.28,29 RT-CGM is associated with an improvement of glycaemic

control28,30-42 and shorter duration of hypoglycaemic episodes.43-45 However, the effect

of conventional RT-CGM (i.e. without automated insulin suspension) specifically in

patients with IAH has not yet been established in a real life setting.46 In most studies

either recent severe hypoglycaemia was among the exclusion criteria30 or the frequency

of severe hypoglycaemia at baseline was not mentioned.38,41,47 Studies performed in

subjects without established impaired awareness of hypoglycaemia already indicate

that RT-CGM might lower the impact of T1DM on daily living29 and increase treatment

satisfaction,48 although the data is limited and inconsistent.49 Since experiencing the

issues mentioned above might especially be the case for patients with T1DM and IAH,

it is likely that RT-CGM will improve the quality of life in these patients.

This trial aims to study a wide range of effects of RT-CGM specifically in T1DM

patients with established IAH, regardless of baseline HbA1c. We hypothesize that the

use of RT-CGM, relative to a control intervention using masked CGM, will result in an

improvement of time spent in normoglycaemia and quality of life in T1DM patients

with IAH.

59

METHODS/DESIGN

Design

This is a two-centre, randomised, crossover trial with a 12-week wash-out period

in between intervention periods (Figure 1). Ethical approval has been granted by the

medical ethical committee of the VU University Medical Center (VUMC).

Weeks since randomisation

Figure 1. General study plan

Period 1 wash out Period 2

RUN IN

RUN IN

RUN INCGM CGM

SMBG SMBG

-6 -2 0 16 28 30 46

Recruitment

Subjects will be recruited from the outpatient clinic of the VUMC, Amsterdam and the

Medical Center Haaglanden (MCH), The Hague, The Netherlands, from outpatient clinics

at affiliated hospitals, but also from regional hospitals and via advertisements in local

newspapers. Responders will be sent written extensive patient information. Written

informed consent will be obtained from the subjects prior to any trial related procedure.

Study population

Fifty-two subjects with T1DM and IAH will be enrolled. Because patients with IAH

often are already in good to moderate control in terms of HbA1c, patients will be included

regardless of their HbA1c.

Ch

apte

r 3

60

Inclusion criteria

• T1DM, diagnosed according to ADA criteria50 regardless duration.

• Use of multiple daily injections of insulin or continuous subcutaneous insulin

infusion. (Participants may not change their treatment method during the

study.)

• Age between 18 and 75 years (inclusive).

• IAH according to the questionnaire by Gold et al.17

• Performing self-monitored blood glucose measurements (SMBG) at least 3/day

or 21/week.

Exclusion criteria

• History of (recent) major renal, liver, or (ischemic) heart disease (including

cardiac conduction disorders).

• Current untreated proliferative diabetic retinopathy.

• Current (treatment for) malignancy.

• Current use of non-selective beta-blockers.

• Current psychiatric disorders, including schizophrenia, bipolar disorder,

anorexia nervosa or bulimia nervosa.

• Substance abuse or alcohol abuse (men >21 units/week, women >14 units/week).

• Current pregnancy or intention to conceive.

• Current use of RT-CGM other than for short term (i.e. diagnostic use or use

shorter than 3 consecutive months).

• Hearing or vision impairment hindering perceiving of glucose display and

alarms, or otherwise incapable of using a (RT-)CGM, in the opinion of the

investigator.

• Poor commandment of the Dutch language or any (mental) disorder that

precludes full understanding of the purpose and instructions of the study.

• Participation in another clinical study.

• Known or suspected allergy to trial product or related products.

Study plan

A detailed overview of study visits, telephone consultations and procedures is given

in Additional file 1. At each visit, a recent history will be taken, including e.g. complaints,

adverse events, changes in medication treatment regime and the occurrence of severe

hypoglycaemia (i.e. requiring assistance of a third party). Weight and vital signs will

be measured at each visit. Outcome measurements will be performed at baseline and

after each 16-week intervention phase (Figure 1).

61

Screening

Hypoglycaemia awareness will be assessed using the questionnaire developed by

Gold et al. and by Clarke et al.17,18 The frequency and severity of severe hypoglycaemia

will be assessed using the self-report questionnaire according to Langan et al.51

Subsequently, a complete medical and socio-economic history will be taken and a

complete physical examination will be performed. Participants will be examined

for the presence of diabetes-related (microvascular) complications. Thus, plasma

creatinine, calculated estimated glomerular filtration rate and albumin-creatinine

ratio in a random urine sample will be measured to evaluate diabetic nephropathy.52,53

The presence/severity of diabetic sensorimotor polyneuropathy will be quantified by

the Modified Toronto Clinical Scoring System.54 For the assessment of vibration sense,

the vibration perception threshold will be quantified using a neurothesiometer.55 The

presence/severity of diabetic retinopathy will be assessed based on the relevant medical

history and a recent (<6 months) evaluation by the patient’s ophthalmologist. In the

absence of this information, patients will be either requested to visit their ophthalmo-

logist prior to study onset or they will be referred for fundus photography.56 Blood will

be collected to ascertain in- and exclusion criteria and for baseline measurement.

Run-in phase

A 5-week run-in phase, starting after inclusion, will be used to allow for study

effects and enable diabetes- and study-related education. All participants will receive

approximately 30 minutes of individual education in the principles of diabetes

management, including basic principles of standardized SMBG, glucose fluctuations,

insulin and carbohydrates, hyper- and hypoglycaemia, impaired awareness of

hypoglycaemia and RT-CGM. In case a participant does not use the technique of

carbohydrate counting, no education on this subject will be given in order to prevent

confounding. Participants will be equipped with a masked CGM device to be worn for

two weeks to gather baseline data and to familiarize them with longer-term wearing

of this type of device. During follow-up visits, there will be an ongoing evaluation of

diabetes self-management and knowledge of diabetes management, as this is part of

standard diabetes care. A test of knowledge will not be used.

Intervention phase

Randomisation

After successful participation in the run-in phase, participants will be randomised,

using block randomisation (allocation 1:1), to the first intervention period. A sealed

Ch

apte

r 3

62

envelope for each participant will be drawn and opened by qualified study staff.

In addition, the block size will be determined by the institutional trial pharmacist.

Study staff and participants will not be blinded for treatment order, as this is an open

intervention study. When allocated to RT-CGM, participants will receive additional

education on how to use RT-CGM and the accompanying specialized software.

Participants will be asked to upload their RT-CGM-data before each follow-up visit.

When allocated to masked CGM, participants will continue to wear their CGM device

from the run-in phase. Participants will be encouraged to wear the assigned study

devices continuously.

Follow-up visits and telephone consultations

During the intervention periods there will be monthly follow-up visits and telephone

consultations involving inquiry after e.g. actual complaints/symptoms, episodes of

(severe) hypoglycaemia, study device use and related technical issues, and medication

using a checklist. During both intervention periods, research physicians will guide

participants in concordance with the ADA Standards of Medical Care in Diabetes57.

The guidance will be equal during both intervention periods. During the visits, therapy

adjustments will be discussed and logged, based on RT-CGM-data in the RT-CGM group

or SMBG-data in the masked CGM group. No treatment or insulin titration protocol

will be used, neither will SMBG use be standardised, in order to avoid additional

interventions.

Wash-out phase

After the first intervention phase, participants will enter a 12-week wash-out phase,

during which they will only receive two-weekly telephone consultations for taking

recent histories and monitoring of potential adverse events. At the end of the wash-out

period, general diabetes and CGM education will be given similar to the first run-in

phase. Also, participants will start to wear masked CGM again for two weeks to gather

baseline data for the second intervention period.

Endpoints

Primary endpoint

The mean difference in time spent in normoglycaemia (interstitial glucose range

>3.9 mmol/L - ≤10.0 mmol/L), expressed as hours/day between the two intervention

periods. The specific ranges of normoglycaemia are based on the definition of the ADA

and comparing literature.12,45

63

Secondary endpoints

• (Diabetes-specific) markers of quality of life, as assessed with questionnaires

covering: diabetes-related emotional distress (PAID-5 [Cronbach’s α = 0.86]58),

fear of hypoglycaemia (HFS [Cronbach’s α = 0.94]59,60), diabetes self-efficacy

(CIDS [Cronbach’s α = 0.94]61), generic health-status (EQ5D [Cronbach’s α =

0.73]62,63) and emotional well-being WHO [Cronbach’s α = 0.82]64-66).

• Other glycaemic variables, including HbA1c, time spent in hypo- and hypergly-

caemia ranges, frequency of severe (i.e. requiring assistance of a third party)

and mild (i.e. self-treated) hypoglycaemia (assessed by analysing (RT-)CGM

data) and duration of hypoglycaemia.

• Changes in hypoglycaemia awareness score according to Gold et al.17

• Glycaemic variability. Glucose variability (defined as inter-day and intraday

glycaemic variability), will be calculated as Mean Of Daily Differences and

Continuous Overall Net Glycaemic Action, using all (RT-)CGM values during the

16 week intervention period.67,68

Exploratory endpoints

• Autonomic nervous system function. The functioning of the autonomic nervous

system will be indirectly assessed by analysing heart rate variability and blood

pressure changes during four of the five Ewing’s standardized cardiovascular

reflex tests: the Valsalva ratio, the 30:15 ratio on standing up, the maximum-

minimum heart rate during deep breathing and postural blood pressure

change.69-72 The heart-rate variability will be assessed using non-invasive,

automated beat-to-beat blood pressure and ECG recordings (Nexfin®, BM Eye,

Amsterdam, the Netherlands). that will be analysed by dedicated software,

automatically calculating HRV indices for both the time-domain and frequen-

cy-domain variables. Quality control will be performed by research staff by

visual inspection of the data. Data will be excluded from analysis if >5% of

the measured beats are extrasystoles. No agreement exists on the number

of abnormal cardiovascular tests required to reach the diagnosis of cardio-

vascular autonomic neuropathy.72 However, an abnormality of more than one

test, on more than one occasion, is indicative of (cardiovascular) autonomic

dysfunction.72,73

• Duration of sensor wear.

• Changes in hypoglycaemia awareness score according to Clarke et al.18

• Satisfaction with use of RT-CGM, assessed by the CGM-SAT questionnaire.74

Ch

apte

r 3

64

Study devices

The MiniMed Paradigm® Veo™ System (Medtronic, Northridge, CA) will be used

as RT-CGM device. The system consists of a continuous Enlite™ glucose sensor, a

MiniLinkTM transmitter and the Paradigm® Veo™ System.75,76 The sensor is a membra-

ne-covered enzyme coated electrode placed through the skin into the subcutaneous

space using an auto-insertion device. This sensor has to be changed every 6 days . The

wireless MiniLinkTM transmitter is a small rechargeable device that is connected to

the glucose sensor and sends glucose data wirelessly to the monitor every 5 minutes,

24 hours a day. The monitor shows real-time glucose measurements and provides

information regarding changes in glucose levels (trends). Calibration is required every

12 hours. The system allows to set alarms, i.e. alarms for low and high glucose limits,

as agreed with the health-care provider, at which the system should alert the patient

whenever glucose values are approximating or exceeding these pre-set values. The

low-glucose limit during this trial will be pre-set at 4,5 mmol/L and cannot be lowered

in order to prevent hypoglycaemia. Also, accuracy of glucose sensors decreases in

hypoglycaemic ranges.77,78 Low-glucose suspension function will not be used. Finally,

the system shows participants their real-time glucose values and patterns, and allows

for adjustment of treatment and/or lifestyle accordingly.

The masked CGM device that will be used is the iPro™2 Continuous Glucose Monitor

(Medtronic, Northridge, CA), which also uses the Enlite™ glucose sensor.75,76 Masked

CGM- and SMBG-data must be uploaded every week for the purpose of (retrospective)

calibration of the iPro™2. Participants will be blinded to the CGM-data. The investigators

will review the CGM-data for quality based on duration of sensor wear, number of

sensor values, accuracy (mean absolute difference), number of valid calibrations.

In case of missing data due to low quality CMG-data, the intervention phase will be

extended.

Statistical considerations

Sample size calculation

Since current RT-CGM technology may show relatively poor performance in the

hypoglycaemic range,79 we chose mean time (hours/day) spent in normoglycaemia, i.e.

glucose values ranging >3.9 - ≤10.0 mmol/L, during the study period as our primary

end-point. A previously published intervention study using RT-CGM demonstrated a

difference of 1.5 hours in time spent in normoglycaemia between the RT-CGM and

control group.45 In order to detect a difference of 1.5 hours (6.25% from 24 hours) in

time spent in normoglycaemia, assuming a standard deviation of 3.5, alpha-level of

65

0.05, power of 80%, a drop-out rate of 15% and a correlation of 0.5 between repeated

measures (which is due to the cross-over design of the study in which subjects serve as

their own control), a sample size of 52 patients will be needed.

Analysis

Intention to treat analysis will include all randomised subjects, including drop-outs.

Per-protocol analysis will include only subjects who have completed at least the first

four weeks of the first intervention period. The total evaluable sample will consist of

all intention-to-treat subjects who complete both intervention periods according to the

protocol. Evaluation of (RT-)CGM data will not be performed in an assessor-blinded way.

Unless otherwise stated, all statistical tests will be conducted at a 2-tailed significance

level of 0.05. According to their distribution, the various parameters will be expressed

as mean (± SD), median (interquartile range) or number (%). Residual effects will be

ascertained before further analysis will be performed using non-parametric tests.80,81

The primary endpoint, i.e. mean difference in time spent in the normoglycaemic range,

expressed as hours/day, during the RT-CGM versus the masked CGM phases, will be

analysed per month using mixed model analysis, in order to show a possible effect

over time. In the mixed model analysis, time spent in normoglycaemia will be used as

dependent variable. Independent variables will include: treatment arm, interaction

between RT-CGM and CGM, and time in months since start of the intervention. Baseline

CGM data, gathered during the run-in phase, will be implemented as covariate. In case

of carry-over effects and an unequal distribution of treatment method (MDI / CSII) over

the intervention orders (RT-CGM – CGM / CGM – RT-CGM), treatment method will also be

included as covariate. By using mixed model analysis, unequal use of the devices will

not necessitate correction, as long as missing data can be considered to be at random.

Carry-over

To date, little is known about the possible carry-over effects of RT-CGM. Yet such an

effect would seem logical considering the expected educational effect of RT-CGM in

e.g. the ability of participants to recognize patterns in otherwise undetected hypo- or

hyperglycaemic events and adjust insulin therapy accordingly. To minimize carry-over

effects, a washout period of 12 weeks will be implanted. We consider a 12-week period

sufficient for the behavioural modification component to wear off. Additionally, a

12-week time period allows realistic HbA1c changes resulting from patients self-ma-

nagement during that same period to establish a reliable baseline for the second

intervention period.

Ch

apte

r 3

66

DISCUSSION

Although it seems obvious that T1DM patients with IAH could potentially benefit

from RT-CGM, the effect of conventional RT-CGM (without automated insulin

suspension) has not yet been established in an ambulatory setting. A hyperinsulinemic

hypoglycaemic clamp study of Ly et al. showed that 4 weeks of RT-CGM improved

epinephrine responses in young T1DM patients with IAH, suggesting that IAH can be

restored in adolescents by using RT-CGM.82 Furthermore, the first observational study

performed in patients with IAH demonstrated a clear reduction of SH with RT-CGM

use,83 addressing the need for further intervention studies in patients with IAH. A

randomised controlled trial of Little et al. demonstrated that IAH can be restored and

SH prevented using existing technology.46 In contrast to expectations, RT-CGM was

not superior over self-monitoring of blood glucose by finger prick. However, during

this trial, strict insulin titration protocols where used, which may not reflect real-life

diabetes management. Although Ly et al. showed that sensor-augmented insulin pump

therapy with automated insulin suspension reduced the frequency of SH significantly

in T1DM patients with IAH 84, this reduction lost significance when 2 outliers, whose

rates of hypoglycaemia were higher at baseline, were removed from analysis. Hence, it

remains unclear whether T1DM patients with IAH benefit from conventional RT-CGM

in real-life.

The study protocol poses some limitations. Hypoglycaemia awareness will be

assessed using the subjective self-rating questionnaire of Gold et al.17 Objective

assessments of hypoglycaemia awareness, i.e. by using CGM data or by inducing

hypoglycaemia in a laboratory setting, will not be used. Self-reported hypoglycaemia

awareness has shown to be reliable in previous studies17-19,85 and reflects usual clinical

practice. Also, continuous use of RT-CGM is encouraged, but not mandatory, which

could influence the effect of RT-CGM on glycaemia.41 Furthermore, the low alarm

value used in this trial is pre-set at 4,5 mmol/L in order to prevent hypoglycaemia.

In clinical practice however, alarm values should be evaluated and set individually.

Moreover, the decrease in sensor accuracy in hypoglycaemic ranges could compromise

the analysis of glycaemia.77,78 Finally, participants already using an insulin pump are

offered a second device, which could diminish compliance with regards to logging

SMBG values, meals, insulin and activities. This could affect the value of interpreting

the RT-CGM-data retrospectively.

67

With this study, we expect to gain a better understanding of the potential benefits

of RT-CGM technology in a sample of patients with T1DM complicated by impaired

hypoglycaemia awareness. Also, the study is likely to yield clinically relevant

information to help further improve targeted intervention strategies to help these

patients improve their diabetes self-management and quality of life.

Ch

apte

r 3

68

1. Forlenza GP, Rewers M. The epidemic

of type 1 diabetes: what is it telling

us? Curr Opin Endocrinol Diabetes

Obes 2011; 18: 248-51.







diabetes mellitus. The Diabetes

Control and Complications Trial

Research Group. N Engl J Med 1993;

329: 977-86.





J Med 2005; 353: 2643-53.

4. MacLeod KM, Gold AE, Frier

BM. Frequency, severity and

symptomatology of hypoglycaemia:

a comparative trial of human and

porcine insulins in type 1 diabetic

patients. Diabet Med 1995; 12:

134-41.






Metab Res Rev 2004; 20: 479-86.

6. Pramming S, Thorsteinsson B,

Bendtson I, Binder C. Symptomatic

hypoglycaemia in 411 type 1 diabetic

patients. Diabet Med 1991; 8: 217-22.

7. Donnelly LA, Morris AD, Frier BM,

et al. Frequency and predictors

of hypoglycaemia in Type 1 and

insulin-treated Type 2 diabetes: a

population-based study. Diabet Med

2005; 22: 749-55.

8. MacLeod KM, Hepburn DA, Frier BM.

Frequency and morbidity of severe

hypoglycaemia in insulin-treated

diabetic patients. Diabet Med 1993;

10: 238-45.

9. ter Braak EW, Appelman AM, van de

Laak M, Stolk RP, van Haeften TW,

Erkelens DW. Clinical characteristics

of type 1 diabetic patients with

and without severe hypoglycemia.

Diabetes Care 2000; 23: 1467-71.



management of Type I and Type

II diabetes. Diabetologia 2002; 45:

937-48.

11. Cryer PE, Davis SN, Shamoon H.

Hypoglycemia in diabetes. Diabetes

Care 2003; 26: 1902-12.

12. Workgroup on Hypoglycemia ADA.





Diabetes Care 2005; 28: 1245-9.


REFERENCES

69



2008; 24: 87-92.




2014.

15. Hepburn DA, Patrick AW, Brash HM,

Thomson I, Frier BM. Hypoglycaemia

unawareness in type 1 diabetes: a

lower plasma glucose is required

to stimulate sympatho-adrenal

activation. Diabet Med 1991; 8:

934-45.




2014; 10: 711-22.






17: 697-703.







Care 1995; 18: 517-22.

19. Geddes J, Wright RJ, Zammitt NN,

Deary IJ, Frier BM. An evaluation

of methods of assessing impaired

awareness of hypoglycemia in type

1 diabetes. Diabetes Care 2007; 30:

1868-70.





Med 2008; 25: 501-4.








young adults. Diabetes Care 2008; 31:

922-6.





since 1989. Diabetologia 2007; 50:

2439-42.




Med 2014; 371: 1972-82.






Care 2006; 29: 44-50.






Diabetes Technol Ther 2003; 5:

769-79.

Ch

apte

r 3

70






Technol 2009; 3: 1207-14.

27. Weinstein RL, Schwartz SL, Brazg RL,

Bugler JR, Peyser TA, McGarraugh

GV. Accuracy of the 5-day FreeStyle

Navigator Continuous Glucose

Monitoring System: comparison

with frequent laboratory reference

measurements. Diabetes Care 2007;

30: 1125-30.







105-13.

29. Geelhoed-Duijvestijn PH, Box F,

Diamant M, Otoide-Vree M, Niel

van J. Real time continuous glucose

monitoring: how does it work in

real life? Br J Diab Vasc Dis 2011; 11:

145-9.





Med 2010; 363: 311-20.


Ahmann A, et al. Sensor-augmented

pump therapy for A1C reduction

(STAR 3) study: results from the

6-month continuation phase.

Diabetes Care 2011; 34: 2403-5.

32. Bode B, Beck RW, Xing D, et al.

Sustained benefit of continuous

glucose monitoring on A1C, glucose

profiles, and hypoglycemia in adults

with type 1 diabetes. Diabetes Care

2009; 32: 2047-9.

33. Danne T, de Valk HW, Kracht T, et

al. Reducing glycaemic variability

in type 1 diabetes self-management

with a continuous glucose

monitoring system based on wired

enzyme technology. Diabetologia

2009; 52: 1496-503.






2006; 29: 2730-2.

35. Garg SK, Schwartz S, Edelman SV.

Improved glucose excursions using

an implantable real-time continuous

glucose sensor in adults with type


734-8.

36. Hirsch IB, Abelseth J, Bode BW,

et al. Sensor-augmented insulin

pump therapy: results of the first

randomized treat-to-target study.


377-83.

37. Ludvigsson J, Hanas R. Continuous

subcutaneous glucose monitoring

improved metabolic control in

pediatric patients with type 1

71

diabetes: a controlled crossover

study. Pediatrics 2003; 111: 933-8.

38. Raccah D, Sulmont V, Reznik Y, et

al. Incremental value of continuous

glucose monitoring when starting

pump therapy in patients with

poorly controlled type 1 diabetes: the

RealTrend study. Diabetes Care 2009;

32: 2245-50.

39. Radermecker RP, Saint RA, Scheen

AJ, Bringer J, Renard E. Continuous

glucose monitoring reduces both

hypoglycaemia and HbA1c in

hypoglycaemia-prone type 1 diabetic

patients treated with a portable

pump. Diabetes Metab 2010; 36:

409-13.

40. Rodbard D, Bailey T, Jovanovic

L, Zisser H, Kaplan R, Garg SK.

Improved quality of glycemic control

and reduced glycemic variability

with use of continuous glucose

monitoring. Diabetes Technol Ther

2009; 11: 717-23.




diabetes. N Engl J Med 2008; 359:

1464-76.






controlled trial. Diabet Med 2011; 28:

1158-67.

43. Davey RJ, Jones TW, Fournier

PA. Effect of short-term use of a

continuous glucose monitoring

system with a real-time glucose

display and a low glucose alarm

on incidence and duration of

hypoglycemia in a home setting in

type 1 diabetes mellitus. J Diabetes

Sci Technol 2010; 4: 1457-64.

44. Beck RW, Hirsch IB, Laffel L, et al.

The effect of continuous glucose

monitoring in well-controlled type


1378-83.






795-800.


E, et al. Recovery of Hypoglycemia

Awareness in Long-standing Type

1 Diabetes: A Multicenter 2 x 2

Factorial Randomized Controlled

Trial Comparing Insulin Pump

With Multiple Daily Injections and

Continuous With Conventional

Glucose Self-monitoring


37: 2114-22.

47. Deiss D, Hartmann R, Schmidt

J, Kordonouri O. Results of a

randomised controlled cross-over

trial on the effect of continuous

subcutaneous glucose monitoring

(CGMS) on glycaemic control in

children and adolescents with type

Ch

apte

r 3

72

1 diabetes. Exp Clin Endocrinol

Diabetes 2006; 114: 63-7.

48. Tansey M, Laffel L, Cheng J, et al.

Satisfaction with continuous glucose

monitoring in adults and youths with

Type 1 diabetes. Diabet Med 2011; 28:

1118-22.

49. Davey RJ, Stevens K, Jones TW,

Fournier PA. The effect of short-term

use of the Guardian RT continuous

glucose monitoring system on fear

of hypoglycaemia in patients with

type 1 diabetes mellitus. Prim Care

Diabetes 2011.

50. Diagnosis and classification of


2012; 35 Suppl 1: S64-S71.

51. Langan SJ, Deary IJ, Hepburn DA,

Frier BM. Cumulative cognitive

impairment following recurrent

severe hypoglycaemia in adult

patients with insulin-treated diabetes

mellitus. Diabetologia 1991; 34:

337-44.

52. Nathan DM, Rosenbaum C,

Protasowicki VD. Single-void urine

samples can be used to estimate

quantitative microalbuminuria.

Diabetes Care 1987; 10: 414-8.

53. Mogensen CE, Keane WF, Bennett

PH, et al. Prevention of diabetic renal

disease with special reference to

microalbuminuria. Lancet 1995; 346:

1080-4.

54. Bril V, Tomioka S, Buchanan RA,

Perkins BA. Reliability and validity

of the modified Toronto Clinical

Neuropathy Score in diabetic

sensorimotor polyneuropathy. Diabet

Med 2009; 26: 240-6.

55. Martin CL, Waberski BH, Pop-Busui

R, et al. Vibration perception

threshold as a measure of distal

symmetrical peripheral neuropathy

in type 1 diabetes: results from the

DCCT/EDIC study. Diabetes Care 2010;

33: 2635-41.

56. Aldington SJ, Kohner EM, Meuer

S, Klein R, Sjolie AK. Methodology

for retinal photography and

assessment of diabetic retinopathy:

the EURODIAB IDDM complications

study. Diabetologia 1995; 38: 437-44.

57. ADA Guidelines 2012. Standards

of medical care in diabetes--2012.


S11-S63.

58. McGuire BE, Morrison TG, Hermanns

N, et al. Short-form measures of

diabetes-related emotional distress:

the Problem Areas in Diabetes Scale

(PAID)-5 and PAID-1. Diabetologia

2010; 53: 66-9.

59. Cox DJ, Irvine A, Gonder-Frederick

L, Nowacek G, Butterfield J. Fear

of hypoglycemia: quantification,

validation, and utilization. Diabetes

Care 1987; 10: 617-21.

60. Gonder-Frederick LA, Schmidt

KM, Vajda KA, et al. Psychometric

properties of the hypoglycemia

fear survey-ii for adults with type


801-6.

73

61. Van Der Ven NC, Weinger K, Yi J,

et al. The confidence in diabetes

self-care scale: psychometric

properties of a new measure of

diabetes-specific self-efficacy in

Dutch and US patients with type 1

diabetes. Diabetes Care 2003; 26:

713-8.

62. Linde L, Sorensen J, Ostergaard M,

Horslev-Petersen K, Hetland ML.

Health-related quality of life: validity,

reliability, and responsiveness

of SF-36, 15D, EQ-5D [corrected]

RAQoL, and HAQ in patients with

rheumatoid arthritis. J Rheumatol

2008; 35: 1528-37.

63. Savoia E, Fantini MP, Pandolfi PP,

Dallolio L, Collina N. Assessing the

construct validity of the Italian

version of the EQ-5D: preliminary

results from a cross-sectional study

in North Italy. Health Qual Life

Outcomes 2006; 4: 47.

64. Awata S, Bech P, Yoshida S, et al.

Reliability and validity of the

Japanese version of the World Health

Organization-Five Well-Being Index

in the context of detecting depression

in diabetic patients. Psychiatry Clin

Neurosci 2007; 61: 112-9.

65. de Wit M, Pouwer F, Gemke RJ,

Delemarre-van de Waal HA,

Snoek FJ. Validation of the WHO-5

Well-Being Index in adolescents with


30: 2003-6.

66. Lowe B, Spitzer RL, Grafe K, et

al. Comparative validity of three

screening questionnaires for DSM-IV

depressive disorders and physicians'

diagnoses. J Affect Disord 2004; 78:

131-40.

67. McDonnell CM, Donath SM, Vidmar

SI, Werther GA, Cameron FJ. A novel

approach to continuous glucose

analysis utilizing glycemic variation.


253-63.

68. Molnar GD, Taylor WF, Ho MM.

Day-to-day variation of continuously

monitored glycaemia: a further

measure of diabetic instability.

Diabetologia 1972; 8: 342-8.

69. Jarrin DC, McGrath JJ, Giovanniello S,

Poirier P, Lambert M. Measurement

fidelity of heart rate variability

signal processing: The devil is in the

details. Int J Psychophysiol 2012.

70. Spallone V, Ziegler D, Freeman R,

et al. Cardiovascular autonomic

neuropathy in diabetes: clinical

impact, assessment, diagnosis, and

management. Diabetes Metab Res

Rev 2011.

71. Ewing DJ, Martyn CN, Young

RJ, Clarke BF. The value of

cardiovascular autonomic function

tests: 10 years experience in


491-8.

72. Spallone V, Bellavere F, Scionti L, et

al. Recommendations for the use of

cardiovascular tests in diagnosing

diabetic autonomic neuropathy. Nutr

Ch

apte

r 3

74

Metab Cardiovasc Dis 2011; 21: 69-78.

73. Consensus statement: Report and

recommendations of the San Antonio

conference on diabetic neuropathy.


American Academy of Neurology.

Diabetes Care 1988; 11: 592-7.

74. Diabetes Research in Children

Network (DirecNet) Study Group.

Youth and parent satisfaction with

clinical use of the GlucoWatch G2

Biographer in the management of

pediatric type 1 diabetes. Diabetes

Care 2005; 28: 1929-35.






Diabetes Obes Metab 2015; 17: 343-9.

76. Keenan DB, Cartaya R, Mastrototaro

JJ. Accuracy of a new real-time

continuous glucose monitoring

algorithm. J Diabetes Sci Technol

2010; 4: 111-8.

77. Clarke WL, Anderson S, Farhy L, et

al. Evaluating the clinical accuracy

of two continuous glucose sensors

using continuous glucose-error grid

analysis. Diabetes Care 2005; 28:

2412-7.

78. Kovatchev B, Anderson S,

Heinemann L, Clarke W. Comparison

of the numerical and clinical

accuracy of four continuous glucose

monitors. Diabetes Care 2008; 31:

1160-4.

79. McGowan K, Thomas W, Moran A.

Spurious reporting of nocturnal

hypoglycemia by CGMS in patients

with tightly controlled type 1


1499-503.

80. Brown BW, Jr. The crossover

experiment for clinical trials.

Biometrics 1980; 36: 69-79.

81. Koch GG. The use of non-parametric

methods in the statistical analysis of

the two-period change-over design.

Biometrics 1972; 28: 577-84.








50-2.







36: 4160-2.








clinical trial. JAMA 2013; 310: 1240-7.

75


S, Thorsteinsson B. Recall of severe

hypoglycaemia and self-estimated

state of awareness in type 1 diabetes.

Diabetes Metab Res Rev 2003; 19:

232-40.

Ch

apte

r 3

76

Visit (V) or telephone call (TC) V1 TC1 V2 V3 V4 TC2 V5 TC3 V6 TC4 V7 TC5 V8 TC6 TC7 TC8 TC9 TC10 V9 V10 TC11 V11 TC12 V12 TC13 V13 TC14 V14

Weeks relative to randomization -6 -5 -2 -1 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

Screening x

Informed consent x

IAH questionnaires (a) x x x x

Assessment of frequency of SH (b) x

Inform about eligibility status x

Medical history x

Recent/actual history x x x x x x x x x x x x x x x x x x x x x x x x x x x

Vital sign/anthropometrics (c) x x x x x x x x x x x x x x

Physical examination x x x x x x x x x x x x x x

Global diabetes education x x

CGM education x x

Diary instructions x x

Sensor change at research unit x

Downloading baseline CGM data x x x

Randomisation x

RT-CGM education x x

Review of RT-CGM data x x x x x x x x

CGM-SAT questionnaire (d) x x

HbA1c x x x x

Lipids (e) x x

Creatinine x x

Haemoglobulin x x

Thyroid stimulating hormone x x

Micro-albuminuria x x

Human chorionic gonadotropin (f) x

Quality of life questionnaires (g) x x x x

ANS function tests (h) x x x x

IAH: impaired awareness of hypoglycaemia; SH: severe hypoglycaemia; CGM: continuous glucose monitoring; RT-CGM: real-time continuous glucose monitoring; ANS: autonomic nervous systema) Hypoglycaemia awareness scores according to Gold et al. and Clarke et al. b) Assessment of frequency of SH according to Langan et al. c) Height will be measured at screening. Blood pressure, heart rate and body weight will be measured at indicated visitsd) Questionnaire for satisfaction with use of CGMe) Including HDL, LDL, triglycerides and total cholesterolf) If the participant is a female in the reproductive ageg) Including PAID-5, HFS-2, CIDS, EQ-5D and WHO-5

Additional file. Table S1 Overview of visits, telephone consultations and outcome assessments

77

Visit (V) or telephone call (TC) V1 TC1 V2 V3 V4 TC2 V5 TC3 V6 TC4 V7 TC5 V8 TC6 TC7 TC8 TC9 TC10 V9 V10 TC11 V11 TC12 V12 TC13 V13 TC14 V14

Weeks relative to randomization -6 -5 -2 -1 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

Screening x

Informed consent x

IAH questionnaires (a) x x x x

Assessment of frequency of SH (b) x

Inform about eligibility status x

Medical history x

Recent/actual history x x x x x x x x x x x x x x x x x x x x x x x x x x x

Vital sign/anthropometrics (c) x x x x x x x x x x x x x x

Physical examination x x x x x x x x x x x x x x

Global diabetes education x x

CGM education x x

Diary instructions x x

Sensor change at research unit x

Downloading baseline CGM data x x x

Randomisation x

RT-CGM education x x

Review of RT-CGM data x x x x x x x x

CGM-SAT questionnaire (d) x x

HbA1c x x x x

Lipids (e) x x

Creatinine x x

Haemoglobulin x x

Thyroid stimulating hormone x x

Micro-albuminuria x x

Human chorionic gonadotropin (f) x

Quality of life questionnaires (g) x x x x

ANS function tests (h) x x x x

IAH: impaired awareness of hypoglycaemia; SH: severe hypoglycaemia; CGM: continuous glucose monitoring; RT-CGM: real-time continuous glucose monitoring; ANS: autonomic nervous systema) Hypoglycaemia awareness scores according to Gold et al. and Clarke et al. b) Assessment of frequency of SH according to Langan et al. c) Height will be measured at screening. Blood pressure, heart rate and body weight will be measured at indicated visitsd) Questionnaire for satisfaction with use of CGMe) Including HDL, LDL, triglycerides and total cholesterolf) If the participant is a female in the reproductive ageg) Including PAID-5, HFS-2, CIDS, EQ-5D and WHO-5

Additional file. Table S1 Overview of visits, telephone consultations and outcome assessments

The Lancet Diabetes & Endocrinology 2016 Nov;4(11):893-902

Cornelis A J van Beers, J Hans DeVries, Susanne J Kleijer, Mark M Smits

Petronella H L M Geelhoed-Duijvestijn, Mark H H Kramer

Michaela Diamant, Frank J Snoek, Erik H Serné

4

Continuous glucose

monitoring for type 1 diabetes

patients with impaired


(IN CONTROL):

a randomised, open-label,

crossover trial

Ch

apte

r 4

80

SUMMARY

Background and aims

Patients with type 1 diabetes who have impaired awareness of hypoglycaemia

have a three to six times increased risk of severe hypoglycaemia. We aimed to assess

whether continuous glucose monitoring (CGM) improves glycaemia and prevents

severe hypoglycaemia compared with self-monitoring of blood glucose (SMBG) in this

high-risk population.

Methods

We did a randomised, open-label, crossover trial (IN CONTROL) at two medical

centres in the Netherlands. Eligible participants were patients diagnosed with type

1 diabetes according to American Diabetes Association criteria, aged 18–75 years,

with impaired awareness of hypoglycaemia as confirmed by a Gold score of at least

4, and treated with either continuous subcutaneous insulin infusion or multiple

daily insulin injections and doing at least three SMBG measurements per day. After

screening, re-education about diabetes management, and a 6-week run-in phase (to

obtain baseline CGM data), we randomly assigned patients (1:1) with a computer-ge-

nerated allocation sequence (block size of four) to either 16 weeks of CGM followed by

12 weeks of washout and 16 weeks of SMBG, or 16 weeks of SMBG followed by 12 weeks

of washout and 16 weeks of CGM (where the SMBG phase was the control). During

the CGM phase, patients used a real-time CGM system consisting of a Paradigm Veo

system with a MiniLink transmitter and an Enlite glucose sensor (Medtronic, CA, USA).

During the SMBG phase, patients were equipped with a masked CGM device, consisting

of an iPro 2 continuous glucose monitor and an Enlite glucose sensor, which does not

display real-time glucose values. The number of SMBG measurements per day and

SMBG systems were not standardised between patients, to mimic real-life conditions.

During both intervention periods, patients attended follow-up visits at the centres

each month and had telephone consultations 2 weeks after each visit inquiring about

adverse events, episodes of hypoglycaemia, etc. The primary endpoint was the mean

difference in percentage of time spent in normoglycaemia (4–10 mmol/L) over the total

intervention periods, analysed on an intention-to-treat basis. Severe hypoglycaemia

(requiring third party assistance) was a secondary endpoint. This trial is registered

with ClinicalTrials.gov, number NCT01787903.

81

Findings

Between March 4, 2013, and Feb 9, 2015, we recruited and randomly assigned 52

patients to either the CGM–SMBG sequence (n=26) or the SMBG–CGM sequence (n=26).

The last patient visit was on March 21, 2016. Time spent in normoglycaemia was

higher during CGM than during SMBG: 65·0% (95% CI 62·8–67·3) versus 55·4% (95% CI

53·1–57·7; mean difference 9·6%, 95% CI 8·0–11·2; p<0·0001), with reductions in both

time spent in hypoglycaemia (ie, blood glucose ≤3·9 mmol/L [6·8% vs 11·4%, mean

difference 4·7%, 95% CI 3·4–5·9; p<0·0001]) and time spent in hyperglycaemia (ie, blood

glucose >10 mmol/L [28·2% vs 33·2%, mean difference 5·0%, 95% CI 3·1–6·9; p<0·0001]).

During CGM, the number of severe hypoglycaemic events was lower (14 events vs

34 events, p=0·033). Five serious adverse events other than severe hypoglycaemia

occurred during the trial, but all were deemed unrelated to the trial intervention.

Additionally, no mild to moderate adverse events were related to the trial intervention.

Interpretation

CGM increased time spent in normoglycaemia and reduced severe hypoglycaemia

in patients with type 1 diabetes and impaired awareness of hypoglycaemia, compared

with SMBG. Our results support the concept of using CGM in this high-risk population.

Ch

apte

r 4

82

Research in context

Evidence before this study

We searched the PubMed database up to May 9, 2016, using the search

terms “continuous glucose monitoring”, “sensor-augmented pump therapy”,

“low-glucose insulin suspension”, predictive low-glucose suspension”, “automated

insulin pump suspension”, “threshold insulin pump interruption”, and “diabetes

mellitus, type 1” for full reports of observational trials, randomised controlled

trials and systematic reviews that investigated the effect of continuous glucose

monitoring (CGM) on glycaemia in patients with type 1 diabetes and impaired

awareness of hypoglycaemia. Our search identified one observational study and

two randomised controlled trials. Findings from the observational study showed

a reduction of severe hypoglycaemia in patients with impaired awareness of

hypoglycaemia, reinforcing the need for randomised studies in patients with

such impaired awareness. Investigators of one of the randomised trials reported

improved hypoglycaemia awareness and glycaemic control from baseline to

endpoint (24 weeks), which did not seem related to use of CGM, but was rather

attributed to extensive interventions including weekly contact, monthly follow-up

visits, and use of a bolus calculator to determine the insulin dose, whether or not

an insulin pump was used. Moreover, sensors were used for a median of 57% of

the time; only 17 of the 42 individuals achieved the 80% sensor usage threshold,

which is often considered the frequency required for meaningful benefit. Findings

from the second randomised controlled trial, which used CGM with low-glucose

suspend, showed a reduction in severe hypoglycaemia in patients with impaired

awareness of hypoglycaemia, but the population studied was quite young (mean

age 18·6 years) and the reduction of severe hypoglycaemia lost significance when

two outliers in the youngest age groups were excluded from the analysis. Since

most patients with impaired awareness of hypoglycaemia are usually older than

40 years and have more than 25 years of diabetes duration, whether CGM adds any

benefit (such as less hypoglycaemia and improved glycaemic control) in patients

with impaired awareness of hypoglycaemia is still unknown.

Added value of this study

We report the findings from our randomised, crossover trial assessing the

effect of CGM without low-glucose suspend on glycaemic control in adult patients

with type 1 diabetes affected by impaired awareness of hypoglycaemia. CGM

improved percentage of time patients spent in normoglycaemia compared with

self-monitoring of blood glucose, by reducing both the percentage of time spent in

83

hypoglycaemia and percentage of time spent in hyperglycaemia. Importantly, the

results also showed CGM reduced severe hypoglycaemia in this typical population

of patients with type 1 diabetes with impaired awareness of hypoglycaemia. In

addition, the absence of an interaction between insulin treatment modality

(multiple daily injections or continuous subcutaneous insulin infusion), and both

the percentage of time spent in normoglycaemia and the proportion of patients

affected by at least one severe hypoglycaemic event are of clinical importance.

Implications of all the available evidence

In earlier trials, CGM did not live up to the expectations of the diabetes community

regarding its ability to reduce severe hypoglycaemia. However, our findings here

support the benefit of CGM, both with and without combining it with continuous

subcutaneous insulin infusion, for improving glycaemic control and diminishing

severe hypoglycaemia in adult patients with type 1 diabetes and impaired

awareness of hypoglycaemia, who are at highest risk of severe hypoglycaemia.

Ch

apte

r 4

84

INTRODUCTION

Maintaining near-normal glucose concentrations lowers the risk of microvascular

and macrovascular complications and reduces mortality in patients with type 1

diabetes.1,2 However, satisfactory glycaemic control is difficult to achieve3 and

hypoglycaemia is a major limiting factor in reaching glycaemic targets.4

Hypoglycaemia has important physical and psychological consequences5,6 and can

even be fatal.7 In adults with type 1 diabetes, the mean incidence of mild hypoglycaemia

is one to two events per patient per week, and the incidence of severe hypoglycaemia

(ie, hypoglycaemia requiring third-party assistance for recovery) is about 0·2 to 3·2

events per patient per year.6 The risk of severe hypoglycaemia increases with increasing

duration of type 1 diabetes. In patients who have had the disease for a long time (>15

years), a prevalence of up to 46% for severe hypoglycaemia, and a mean frequency of

3·2 episodes per patient-year have been reported.8 Recurrent hypoglycaemia induces

defective glucose counter-regulation and impaired awareness of hypo glycaemia.9 This

impaired awareness occurs in roughly 25% of adult patients with type 1 diabetes10 and

renders patients at a three to six times increased risk of severe hypoglycaemia.10,11

Continuous glucose monitoring (CGM) reduces HbA1c without increasing

hypoglycaemia, with the largest effect in patients with the highest HbA1c at baseline.12

Current marketed CGM systems are used as standalone devices, or are connected

to insulin pumps (sensor-augmented pump therapy), with or without a (predicted)

low-glucose suspend feature, which automatically interrupts insulin administration

for up to two hours when glucose concentration falls below a pre-set threshold.13

Findings from an observational study14 have suggested that CGM reduces the risk

of severe hypoglycaemia in patients with type 1 diabetes and impaired awareness of

hypoglycaemia. This finding was supported by results from a randomised controlled

trial using sensor-augmented pump therapy with low-glucose suspension.15 However,

the population studied in the trial was quite young (mean age 18·6 years) and the

reduction of severe hypoglycaemia was not significant when two outliers in the

youngest age groups were excluded from the analysis. Since most patients with

impaired awareness of hypoglycaemia are older than 40 years and have had diabetes

for more than 25 years,16 whether CGM improves glycaemia more than self-monitoring

of blood glucose (SMBG) in a typical adult type 1 diabetes population with impaired

awareness of hypoglycaemia has yet to be determined.17

85

Therefore, the primary objective of this trial was to investigate the effect of CGM

compared with SMBG on glycaemic control in adult patients with type 1 diabetes and

impaired awareness of hypoglycaemia.

Ch

apte

r 4

86

METHODS

Study design and participants

A detailed description of the study protocol has been previously published.18

Briefly, we did a two-centre, randomised, crossover, open-label trial (IN CONTROL) at

the VU University Medical Center (Amsterdam, Netherlands) and the Medical Center

Haaglanden (The Hague, Netherlands). Ethical approval was granted by the medical

ethical committee of the VU University Medical Center. We recruited patients from

the outpatient clinics of both medical centres, and from outpatient clinics at affiliated

hospitals. To be eligible, patients had to be diagnosed with type 1 diabetes (based

on American Diabetes Association [ADA] criteria),19 aged 18–75 years, be treated

with either continuous subcutaneous insulin infusion (CSII) or multiple daily insulin

injections (MDI), be undertaking at least three SMBG measurements per day, and have

impaired awareness of hypoglycaemia as defined by Gold criteria (ie, with a Gold

score ≥4).11 The Gold method has previously been validated in adult patients with type

1 diabetes and is easy to use.11 Patients were excluded if they had a history of renal,

liver, or heart disease, current untreated proliferative diabetic retinopathy, current

malignancy, current use of non-selective β blockers, current psychiatric disorders,

current substance abuse or alcohol abuse, pregnancy, current use of CGM other than

for a short period (3 consecutive months), any hearing or vision impairment that

could hinder perception of the glucose display and alarms, poor command of the

Dutch language or any disorder that precluded full understanding of the purpose

and instructions of the study, participation in another clinical study, and any known

or suspected allergy to trial-related products. We obtained written informed consent

from the patients before any trial related procedures began. In accordance with

the risk assessment used in the Netherlands to establish the need for a data safety

monitoring board, the present study did not need to have a data safety monitoring

board.

Randomisation and masking

After screening and a 6-week run-in phase (including re-education about diabetes

management given 2 weeks before randomisation), we randomly assigned patients

(1:1) using a computer-generated allocation sequence (block size of four) to either 16

weeks of CGM followed by 12 weeks of washout and 16 weeks of SMBG, or 16 weeks of

SMBG followed by 12 weeks of washout and 16 weeks of CGM, where the SMBG phase

was the control. The allocation sequence (CGM–SMBG or SMBG–CGM) was generated

by the institutional trial pharmacist, and masked to the physicians (by use of sealed

envelopes) at the time of randomisation (ensuring low risk of allocation bias). After

randomisation, the sequence was no longer masked for both study physicians (who

87

also assessed outcomes and analysed the data) and patients.

Procedures

The (re)education about diabetes management given to all patients before

randomisation covered the basic principles of SMBG, hyperglycaemia and

hypoglycaemia, glucose fluctuations, insulin and carbohydrates, impaired awareness

of hypoglycaemia, and safe and effective use of CGM. No education about the technique

of carbohydrate counting was given, in case patients did not practise this technique

before enrolment. Patients were equipped with a masked CGM system consisting of an

iPro 2 continuous glucose monitor and an Enlite glucose sensor (Medtronic, Northridge,

CA, USA), for 2 weeks. This masked CGM system does not display real-time CGM data

or glucose trends or allow alarms to be set. The Enlite sensors have a mean absolute

relative difference between sensor and reference values of less than 20%.20,21 Patients

were eligible for randomisation if the maximum number of sensor values per day (288)

for at least 4 days per week had been obtained, three to four valid calibrations per

day had been done, and a daily mean absolute difference less than 18% (in case of a

difference between the highest and the lowest calibration value <5·6 mmol/L) or a daily

mean absolute difference less than 28% (in case of a difference between the highest

and the lowest calibration value ≥5·6 mmol/L) was noted. These cutoff values are used

in our clinical practice, and were based on CGM manufacturers’ advice (Medtronic,

personal communication). In case of low quality or missing CGM data, the run-in

phase was extended until satisfactory CGM data for at least 4 days per week had been

obtained.

During both intervention periods, patients attended monthly follow-up visits

followed by telephone consultations 2 weeks after each follow-up visit, involving inquiry

about adverse events, all episodes of hypoglycaemia including severe episodes, use of

study device and related technical issues, and to check current medication. Treatment

goals were equal in both study periods and in concordance with the ADA Standards

of Medical Care.22 Patients continued using their own blood glucose meters. Therapy

adjustments were made on the basis of CGM data in the CGM phase or SMBG data in

the SMBG phase. No specific educational issues were addressed other than those stated

in the ADA Standards of Medical Care, no treatment or insulin titration protocols were

used, and SMBG was not standardised between patients, to mimic real-life conditions

and avoid additional interventions. After the first intervention period, patients entered

a 12-week washout phase, during which they only received telephone consultations

for taking recent medical history and monitoring of potential adverse events every 2

weeks. At the end of the washout period, the general diabetes education was repeated

and patients wore a masked CGM device again for 2 weeks to gather baseline data for

the second intervention period. At baseline and endpoint of both intervention periods,

Ch

apte

r 4

88

HbA1c was measured by high-performance liquid chromatography and self-reported

hypoglycaemia awareness was assessed with the Gold11 and Clarke23 methods.

The CGM system used during the intervention phase consisted of the Paradigm Veo

system used solely as a monitor with a MiniLink transmitter (Medtronic, Northridge,

CA, USA for both), and the Enlite glucose sensor. CSII-treated patients continued using

their own pump for insulin treatment. The low-glucose limit during this trial was

preset at 4·5 mmol/L and the low-glucose suspension function was not used. CGM data

were uploaded before every follow-up visit. Patients were encouraged to use CGM

continuously, although this use was not mandatory. During the SMBG phase, patients

wore the masked CGM system continuously throughout the intervention phase and

uploaded the masked CGM data each week. Because of frequent issues with uploading

data from the masked CGM device, we assessed the quality of the CGM data and

included these data in the analysis if at least 4 days per week’s worth of satisfactory

CGM data, based on the same criteria as in the run-in phase, were obtained. In case of

low quality or missing CGM data, the intervention phase was extended until at least 2

weeks of satisfactory CGM data in a 4-week period had been obtained.

Outcomes

The primary outcome was the mean difference in the percentage of time that

patients spent in normoglycaemia (4·0–10·0 mmol/L) between CGM and SMBG

calculated over the total intervention periods. In the original protocol, time spent

in range was expressed as h per day, but this was redefined as percentages because

CGM trials most often report this outcome in this way.15,17,18 (The change was included

on April 2, 2015, but no official protocol amendment was made because this change

was not considered substantial.) Secondary endpoints were time spent in normogly-

caemia each month to show an effect over time, severe hypoglycaemia (defined as

a hypoglycaemic event requiring third-party assistance), the percentage of time

patients spent in a hypoglycaemic state (blood glucose ≤3·9 mmol/L) and a hypergly-

caemic state (>10·0 mmol/L), average daily area under the curve (AUC) of 3·9 mmol/L

or less (expressed as mmol/L min), frequency (episodes per week) and duration (min

per episode) of CGM-derived hypo glycaemic episodes (≥three sequential sensor

values ≤3·9 mmol/L), frequency (episode per night) and duration of CGM-derived

hypoglycaemic episodes at night-time (0000–0600 h), and within-day and between-day

glucose variability (calculated as within-day SD of glucose concentration, coefficient of

variation, mean absolute change in glucose concentration, mean of daily differences,

and continuous overall net glycaemic action).24,25 Other secondary endpoints were

baseline and 16-week HbA1c measurements, self-reported hypoglycaemia awareness

(based on Gold11 and Clarke23 methods), diabetes-specific measures of quality of life

(PAID-5, HFS, CIDS, EQ5D, and WHO-5),18 and satisfaction with use of CGM assessed by

89

the CGM-SAT questionnaire.18 We also assessed post hoc the frequency of CGM-derived

hypoglycaemic episodes with cutoffs of less than 3·5 mmol/L and less than 2·8 mmol/L.

Other outcomes specified in our protocol (qualitative analysis of experience with CGM

and function of autonomic nervous system) are beyond the scope of this report and

will be reported elsewhere.

Statistical analysis

Since the results from a published CGM trial26 showed a difference of 1·5 h (6·25%)

in time spent in normoglycaemia between CGM and SMBG, we aimed to detect such a

difference, assuming an SD of 3·5 h, an α of 0·05, a power of 80%, and a correlation of

0·5 between repeated measures. Assuming that about 15% of patients would drop out,

we calculated that a sample size of 52 patients was needed.

We did all statistical tests at a two-tailed significance level of 0·05. We analysed the

primary endpoint in the intention-to-treat population using a linear mixed-model

analysis with the percentage of time spent in normoglycaemia as the dependent

variable, the treatment group (CGM or SMBG) as a factor, and the participant as a

random factor. Because of the crossover design of our trial, we assessed the carryover

effect by including the sequence allocation as a factor in the mixed model. If a carryover

effect was detected (p<0·1), only the first study period was analysed (treating it as a

parallel randomised controlled trial). Additionally, insulin treatment modality (MDI

or CSII) was included as a covariate in the model and a p value for interaction of 0·1

was regarded as significant. The percentage of time spent in normoglycaemia was

also analysed per month by including the time since the start of the trial in months as

a covariate in the mixed model. All other outcomes were also analysed in the intenti-

on-to-treat population. We analysed the outcomes with a Gaussian distribution using

a similar model, and analysed non-Gaussian distributed data using the Wilcoxon

matched-pair signed-rank test.

We analysed the proportion of patients with at least one severe hypoglycaemic

event using a generalised estimating equation (GEE) with a logistic link function. We

preferred to use GEE rather than generalised linear mixed models, because problems

of convergence are more likely to occur when using the linear mixed models method.

We used an exchangeable correlation to account for correlation between repeated

observations for the same patient. We did all analyses with SPSS 22.0 for Windows.

This trial is registered with ClinicalTrial.gov, number NCT01787903.

Role of the funding source

The funders of the study had no role in study design, data collection, data analysis,

data interpretation, or writing of the report. CAJvB, SJK, MMS, PHG-D, and EHS had

Ch

apte

r 4

90

access to the raw data. The corresponding author had full access to all the data in the

study and had final responsibility for the decision to submit for publication.

91

RESULTS

Between March 4, 2013, and Feb 9, 2015, 57 patients attended the screening visit,

and 52 were randomly assigned to either the CGM-to-SMBG sequence (n=26) or to

the SMBG-to-CGM sequence (n=26; figure 1). The first patient was enrolled on March

4, 2013, and the final patient’s last visit was on March 21, 2016. Five patients were

deemed ineligible: two had a Gold score of 3, one had type 2 diabetes, one had a

malignancy, and one was deemed unable to adhere to the study protocol, because he

57 patients assessed for eligibility

52 enrolled

5 ineligible

52 randomised

26 assigned to CGM 26 assigned to SMBG

3 discontinued treatment

3 withdrew consent


2 withdrew consent

23 assigned to SMBG 24 assigned to CGM


1 withdrew consent

26 included in intention-to-treat

analysis

26 included in intention-to-treat

analysis

CGM: Continuous glucose monitoring; SMBG: Self-monitoring of blood glucose.

Figure 1. Trial profile

Ch

apte

r 4

92

could not attend the monthly follow-up visits. After randomisation, six patients (12%)

withdrew early: two discontinued after the CGM period because of motivational issues,

one had personal circumstances necessitating discontinuation, two withdrew because

they could not upload the masked CGM device, and one withdrew because of poor

adherence to CGM (baseline characteristics presented in table 1). 18 patients (35%) of

52 practised the technique of carbohydrate counting.

All 52 randomly assigned patients were included in the primary analysis. Only

Intention-to-treat population (n=52)

Woman 24 (46·2%)

Age (years) 48·6 (11·6)

Weight (kg) 76·9 (14·2)

BMI (kg/m2) 25·0 (3·8)

Diabetes duration (years) 30·5 (18·5-40·8)

HbA1c (%) 7·5 (0·8)

HbA1c (mmol/mol) 58·1 (8·7)

Insulin treatment modality, CSII 23 (44·2%)

Duration of CSII therapy (years) 10·2 (4·7-18·7)

Total daily insulin dose (U/kg) 0·5 (0·4-0·7)

Self-reported daily home glucose-meter readings (no./day) 5·0 (4·0-6·0)

Carbohydrate counting 18 (34·6%)

Retinopathy* 24 (46·2%)

Peripheral neuropathy† 14 (26·9%)

Microalbuminuria‡ 8 (15·4%)

Gold score11 5·4 (0·7)

Clarke score23 5·0 (1·3)

Impaired awareness of hypoglycaemia according to Gold11 and Clarke23

45 (86·5%)

Frequency of severe hypoglycaemia§

More than one episode per week 2 (4%)

More than one episode per month 7 (15%)

4 to 12 episodes per year 9 (20%)

1 to 3 episodes per year 14 (30%)

Less than one episode per year 11 (24%)

Never had severe hypoglycaemia 3 (7%)

Data are mean (SD), median (IQR), or n (%). CSII=continuous subcutaneous insulin infusion. *Based on medical history or

fundus photography. †Based on medical history or physical examination. ‡Based on an increased albumin-tocreatinine

ratio (>2·5 mg/mmol for men and >3·5 mg/mmol for women) or current treatment for microalbuminuria. §46 of 52 parti-

cipants filled out the severe hypoglycaemia questionnaire.

Table 1. Baseline characteristics of the intention-to-treat population

93

one participant had used CGM before, during a marathon more than 6 months before

randomisation. Median sensor use during the CGM period was 89·4% (IQR 80·8–95·5).

Means of 13·0 weeks (SD 1·8) of masked CGM data and 13·6 weeks (SD 1·9) of real-time

CGM data were obtained per patient during the study phases, so differences in the

amount of CGM data between the CGM and SMBG phases in our analyses did not need

to be controlled for.

The percentage of time that patients spent in a normoglycaemic state during the

16-week intervention period was higher during CGM than during SMBG (65·0% [95%

CI 62·8–67·3] vs 55·4% [95% CI 53·1–57·7]; mean difference 9·6%, 95% CI 8·0–11·2;

p<0·0001; table 2, figure 2), and the number of h per day that patients spent in the

normoglycaemic state was higher during CGM. All other outcomes were lower during

CGM than during SMBG (table 2).

week

Datapoints represent mean (SE) percentage of time spent in normoglycaemia (4·0–10·0 mmol/L) during the preceding 4 weeks

in patients allocated to the CGM–SMBG sequence (red line) and SMBG–CGM sequence (blue line). CGM=continuous glucose

monitoring. SMBG=self-monitoring of blood glucose. Datapoints at week 0 and week 30 represent mean (SE) percentage of time

spent in normoglycaemia during preceding 2 run-in weeks. Period 1: red line=CGM, blue line=SMBG. Period 2: red line=SMBG,

blue line=CGM.

Figure 2. Percentage of time spent in normoglycaemia

Tim

e sp

ent

in n

orm

ogly

caem

ia (

%)

0 4 8 12 16 30 34 38 42 46

70

65

60

55

50

0

Period 2Period 1

Ch

apte

r 4

94

CG

MS

MB

GM

ean

dif

fere

nce

(9

5% C

I)p

va

lue

Per

cen

t of

tim

e sp

ent

wit

h g

luco

se le

vel i

n r

ange

4·0-

10 m

mol

/L65

·0 (

62·8

to

67·3

)55

·4 (

53·1

to

57·7

)9·

6 (8

·0 t

o 11

·2)

<0·0

001

≤3·9

mm

ol/L

6·8

(5·2

to

8·3)

11·4

(9·

9 to

13·

0)-4

·7 (

-5·9

to

-3·4

)<0

·000

1

>10

mm

ol/L

28·2

(25

·1 t

o 31

·3)

33·2

(30

·0 t

o 36

·3)

-5·0

(-6

·9 t

o -3

·1)

<0·0

001

Tim

e sp

ent

wit

h g

luco

se le

vel i

n r

ange

(h

ours

/day

)

4·0-

10 m

mol

/L15

·6 (

15·1

to

16·2

)13

·3 (

12·7

to

13·8

)2·

3 (1

·9 t

o 2·

7)<0

·000

1

≤3·9

mm

ol/L

1·6

(1·3

to

2·0)

2·7

(2·4

to

3·1)

-1·1

(-1

·4 t

o -0

·8)

<0·0

001

>10

mm

ol/L

6·8

(6·0

to

7·5)

8·0

(7·2

to

8·7)

-1·2

(-1

·6 t

o -0

·7)

<0·0

001

CG

M-d

eriv

ed h

ypog

lyca

emic

eve

nts

(ev

ents

/wee

k)10

·1 (

8·7

to 1

1·4)

11·1

(9·

8 to

12·

5)-1

·1 (

-2·1

to

-0·1

)0·

028

Du

rati

on o

f C

GM

-der

ived

hyp

ogly

caem

ic e

ven

ts (

min

/ev

ent)

60·7

(54

·9 t

o 66

·4)

98·5

(92

·6 t

o 10

4·3)

-37·

8 (-

44·6

to

-30·

9)<0

·000

1

AU

C ≤

3·9

mm

ol/L

per

24

hou

rs (

mm

ol/L

x m

inu

te)

62·9

(45

·1 t

o 80

·7)

115·

8 (9

7·8

to 1

33·8

)-5

2·9

(-68

·1 t

o -3

7·7)

<0·0

001

Noc

turn

al h

ypog

lyca

emia

(00

00-0

600

hou

rs)

Per

cen

t of

tim

e sp

ent ≤3

·9 m

mol

/L7·

6 (5

·3 t

o 9·

8)13

·3 (

11·0

to

15·5

)-5

·7 (

-8·2

to

-3·2

)<0

·000

1

Nu

mb

er o

f C

GM

-der

ived

hyp

ogly

caem

ic e

ven

ts p

er

nig

ht

0·26

(0·

21 t

o 0·

31)

0·33

(0.

28 t

o 0·

38)

-0·0

7 (-

0.11

to

-0·0

2)0·

003

Du

rati

on o

f C

GM

-der

ived

hyp

ogly

caem

ic e

ven

ts a

t n

igh

t-ti

me

(min

/eve

nt)

78·7

(69

·3 t

o 88

·1)

131·

4 (1

21·9

to

140·

9)-5

2·7

(-62

·7 t

o -4

2·7)

<0·0

001

Mea

n g

luco

se (

mm

ol/L

)8·

3 (8

·0 t

o 8·

6)8·

7 (8

·4 t

o 9·

0)-0

·4 (

-0·6

to

-0·2

)0·

001

Wit

hin

day

SD

of

glu

cose

(m

mol

/L)

2·8

(2·7

to

2·9)

3·3

(3·1

to

3·4)

-0·5

(-0

·6 t

o -0

·4)

<0·0

001

Coe

ffic

ien

t of

var

iati

on o

f gl

uco

se (

%)

Ove

rall

39·5

(38

·2 t

o 40

·8)

46·3

(44

·9 t

o 47

·6)

-6·7

(-8

·0 t

o -5

·5)

<0·0

001

Wit

hin

day

33·5

(32

·4 t

o 34

·6)

38·0

(36

·9 t

o 39

·1)

-4·5

(-5

·5 t

o -3

·6)

<0·0

001

Bet

wee

n d

ays

18·4

(17

·5 t

o 19

·4)

23·1

(22

·2 t

o 24

·1)

-4·7

(-5

·9 t

o -3

·5)

<0·0

001

MA

G (

mm

ol/L

/hou

r)1·

7 (1

·7 t

o 1·

8)1·

8 (1

·7 t

o 1·

9)-0

·1 (

-0·1

to

-0·0

)0·

049

MO

DD

(m

mol

/L)

3·3

(3·1

to

3·5)

4·2

(4·0

to

4·4)

-0·9

(-1

·1 t

o 0·

7)<0

·000

1

95

CO

NG

A1 (m

mol

/L)

1·7

(1·6

to

1·8)

1·8

(1·7

to

1·9)

-0·1

(-0

·2 t

o -0

·0)

0·00

2

Dat

a ar

e m

ean

(95

% C

I). C

GM

: Con

tin

uou

s gl

uco

se m

onit

orin

g; S

MB

G: S

elf-

mon

itor

ing

of b

lood

glu

cose

; AU

C: A

rea

un

der

th

e cu

rve;

SD

: Sta

nd

ard

dev

iati

on; M

AG

: Mea

n a

bso

lute

glu

cose

chan

ge; M

OD

D: m

ean

of

dai

ly d

iffe

ren

ce; C

ON

GA

1: c

onti

nu

ous

over

all n

et g

lyca

emic

act

ion

at

1 h

inte

rval

s.

Ta

ble

2. C

GM

-der

ived

ou

tco

mes

CG

MS

MB

GO

dd

s ra

tio

(9

5% C

I)M

ean

dif

fere

nce

(9

5% C

I)p

va

lue

Sev

ere

Hyp

ogl

yca

emia

(S

H)

Nu

mb

er o

f se

vere

hyp

ogly

caem

ic e

ven

ts14

340·

033*

Pro

por

tion

of

pat

ien

ts w

ith

≥1

SH e

ven

t (n

)0·

19 (

10)

0·35

(18

)0·

48 (

0·2

to 1

·0)†

0·06

2

Hb

A1c

Hb

A1c

, en

dp

oin

t (%

) 7·

3 (7

·1 t

o 7·

5)7·

3 (7

·1 t

o 7·

5)0·

0 (-

0·1

to 0

·2)

0·81

2

Ch

ange

in H

bA

1c f

rom

bas

elin

e (%

)-0

·1 (

-0·2

to

0·1)

-0·1

(-0

·2 t

o 0·

0)-0

·1 (

-0·2

to

0·1)

0·44

9

Sel

f-re

po

rted

hyp

ogl

yca

emia

aw

are

nes

s

Gol

d s

core

,11 e

nd

poi

nt‡

4·6

(4·3

to

5·0)

5·0

(4·6

to

5·4)

-0·4

(-0

·7 t

o 0·

0)0·

035

Ch

ange

in G

old

sco

re f

rom

bas

elin

e -0

·5 (

-0·8

to

-0·1

)-0

·1 (

-0.4

to

0.2)

-0·4

(-0

·8 t

o 0·

0)0·

076

Cla

rke

scor

e,23

en

dp

oin

t‡4·

4 (3

·9 t

o 4·

8)4·

4 (3

·9 t

o 4·

8)0·

0 (-

0·4

to 0

·4)

0·95

3

Ch

ange

in C

lark

e sc

ore

from

bas

elin

e-0

·1 (

-0·5

to

0.3)

-0·4

(-0

·8 t

o 0·

0)-0

·3 (

-0·9

to

0·2)

0·21

6

Dat

a ar

e n

(%

) or

mea

n (

95%

CI)

, un

less

oth

erw

ise

ind

icat

ed. C

GM

=con

tin

uou

s gl

uco

se m

onit

orin

g. S

MB

G=s

elf-

mon

itor

ing

of b

lood

glu

cose

. *R

esu

lt o

f re

late

d-s

amp

les

Wil

coxo

n s

ign

ed-r

ank

test

don

e fo

r 16

-wee

k se

vere

hyp

ogly

caem

ia e

ven

t ra

tes

per

100

pat

ien

t-m

onth

s. †

Res

ult

of

gen

eral

ised

est

imat

ing

equ

atio

n a

nal

ysis

ad

just

ed f

or s

tud

y d

ura

tion

. ‡m

easu

red

aft

er t

he

16-w

eek

inte

rven

tion

ph

ase.

Ta

ble

3. S

ever

e h

ypo

glyc

aem

ia, H

bA

1c, a

nd

sel

f-re

po

rted

hyp

ogl

yca

emia

aw

are

nes

s

Ch

apte

r 4

96

Fewer severe hypoglycaemic events occurred during CGM than with SMBG (table

3, figure 3). During both the CGM and SMBG phases, four severe hypoglycaemic

events occurred resulting in seizure or coma, and one severe hypoglycaemic event

resulted in the patient being admitted to the hospital. Ten patients (19%) had one or

more severe hypoglycaemic event during CGM, compared with 18 (35%) during SMBG

(uncorrected odds ratio [OR] 0·45, 95% CI 0·23–0·87; p=0·018), with no interaction for

insulin treatment modality (CSII vs MDI; p=0·348). However, because of issues with

uploading the masked CGM device, the median duration of the SMBG phase was 18·0

weeks (IQR 16·3–20·9) versus 16·0 weeks (16·0–16·9) in the CGM phase. Correction for

study duration in the GEE model did not substantially affect the point estimate for

the OR (0·48, 0·22–1·04; p=0·062), although the finding was not significant after this

correction was made. After completion of the CGM phase, time spent in the normog-

lycaemic state reverted towards baseline values after the 12-week washout period

(figure 2). The sequence allocation had no effect on the primary endpoint (p=0·548)

and the effect was constant over both study periods (p=0·157). We noted no differences

for the percentage of time that patients spent in a normoglycaemic state between

those on MDI versus CSII (figure 4), or between patients who used or did not use the

technique of carbohydrate counting (p for interaction 0·634, and 0·938, respectively).

One patient who spent less time in normoglycaemia during the CGM phase (52·7%)

compared with the SMBG phase (57·3%) also spent less time in hypoglycaemia during

the CGM phase (4·8%) compared with the SMBG phase (16·5%). The mean HbA1c after

both intervention periods and the change in HbA1c from baseline to endpoint were

equal in the CGM and SMBG groups (table 3).

We noted no relevant differences in self-reported hypoglycaemia awareness scores,

with no relevant between-group differences in 16-week hypoglycaemia awareness

scores or change in hypoglycaemia awareness scores from baseline to endpoint (table

3). No between-group differences were noted in quality of life from scores on the HFS

Behaviour subscale, PAID-5, CIDS, EQ5D, or WHO-5 between the CGM and SMBG phases

(data not shown). Scores on the HFS Worry subscale, trans formed to a 0–100 scale,

were lower after the CGM phase compared with the SMBG phase (32·5 vs 38·9; mean

difference 6·4, 95% CI 1·4–11·4; p=0·014). CGM-SAT scores after the CGM phase were

higher than neutral (3·0 on a 5·0 scale), with a mean score of 3·8 (SD 0·6).

Five serious adverse events other than severe hypoglycaemia occurred during the

trial, but none were deemed related to the study intervention. In the washout phase,

one event each of anaphylactic reaction to eye drops, cerebral contusion, rupture of the

Achilles tendon, and rupture of the biceps tendon occurred; one hospital admission for

erysipelas (not at the CGM insertion site) occurred during the CGM phase. No ketoacidosis

97

Severe hypoglycaemia requiring

third-party assistance

Severe hypoglaemia resulting in

coma or seizure

Severe hypoglycaemia resulting in

admission to hospital

Intervention

CGMSMBG

<2.8 mmol/L<3.5 mmol/L

Biochemical cutoff

<3.5 mmol/L

Biochemical cutoff

≤3.9 mmol/L

*Mean difference 44.0%



20

15

10

5

0

CG

M-d

eriv

ed h

ypoc

glae

mia

(ev

ents

per

wee

k)

40

30

20

10

0

Seve

re h

ypog

lyca

emic

eve

nt

(n)

SMBG

CMG

(A) Bars represent mean (SD) frequency of CGM-derived hypoglycaemic events per week during the total SMBG period (red

bars) and the total CGM period (blue bars), grouped per biochemical cutoff . (B) Number of severe hypoglycaemic events requi-

ring third-party assistance, resulting in coma or seizure, or resulting in admission to hospital are shown for the SMBG period

and the CGM period. SMBG=self-monitoring of blood glucose. CGM=continuous glucose monitoring. *Denotes p<0·05 for the

comparisons between CGM and SMBG per biochemical cutoff . †Result of the related-samples Wilcoxon signed-rank test done

on rates of 16-week severe hypoglycaemic events (requiring third-party assistance) per 100 patient-months (p=0·033).

Figure 3. (A) CGM-derived hypoglycaemia and (B) severe hypoglycaemia

B

A

Ch

apte

r 4

98

occurred during the trial. 11

mild to moderate adverse events

occurred during the CGM phase,

16 mild to moderate adverse

events occurred during the SMBG

phase, and two mild to moderate

adverse events occurred during

the wash-out phase. The mild to

moderate adverse events were

deemed unrelated to the study

intervention and consisted

of adverse events related to

the musculoskeletal system

(CGM phase, n=2; SMBG phase,

n=6), urinary tract infection

(CGM phase, n=2), dermal

infection(CGM phase, n=2; SMBG

phase, n=1; wash-out, n=1),

gastrointestinal infection (CGM

phase, n=1; SMBG phase, n=3),

dermal burn (CGM phase, n=1),

fever for less than 1 week (CGM

phase, n=2; SMBG phase, n=3),

excision of lipoma (CGM phase,

n=1), dyspnoea (washout, n=1),

periodontitis (SMBG phase, n=1),

infection of the upper respiratory

tract (SMBG phase, n=1), and

glaucoma (SMBG phase, n=1).

SMBG CGMIntervention

85

80

75

70

65

60

55

50

45

40

0

CSII

MDI

Tim

e sp

ent

in n

orm

ogly

caem

ia (

%)

Data are the mean percentage of time spent in normoglycaemia (4·0–10·0

mmol/L) for each patient for the total SMBG period, which is connected

to the percentage of time spent in normoglycaemia for the total CGM pe-

riod. Blue lines represent patients treated with CSII. Red lines represent

patients treated with MDI. Thicker lines connect the mean percentage of

time spent in normoglycaemia in the total SMBG period and the total CGM

period for all patients treated with CSII (blue line) and for all patients tre-

ated with MDI (red line). CSII=continuous subcutaneous insulin infusion.

MDI=multiple daily injections. SMBG=self-monitoring of blood glucose.

CGM=continuous glucose monitoring.

Figure 4. Percentage of time spent in normoglycaemia for each patient

99

DISCUSSION

The results from our randomised controlled crossover trial in adult patients with

type 1 diabetes and impaired awareness of hypoglycaemia showed that a 16-week

intervention with CGM (without low-glucose suspension) significantly improved time

that the patients spent in a normoglycaemic state, with less time spent in hypoglycaemia

and hyperglycaemia, compared with SMBG. Additionally, CGM decreased the frequency

of severe hypoglycaemic events in this high-risk population, and produced less glucose

variability, but without a change in HbA1c. Our findings suggest that CGM is a valuable

tool for the treatment of adult patients with type 1 diabetes and impaired awareness

of hypoglycaemia. Our reported differences in time spent in normoglycaemia between

SMBG and CGM are similar to the difference between patients using CGM less than 50%

and 50% or more of the time, as reported in another intervention trial17 in patients

with impaired awareness of hypoglycaemia. The findings of our trial lend support to

the belief that CGM does not have an effect beyond the actual intervention, because

withdrawal of CGM resulted in a reversal of the time spent in the normoglycaemic state

to baseline values after 12 weeks.27

In addition to reducing the frequency of CGM-derived hypoglycaemia and severe

hypoglycaemia, the 16-week CGM phase also saw a smaller proportion of patients

affected by severe hypoglycaemia than did the SMBG phase. Importantly, this effect

occurred without increasing HbA1c, which is often the price paid when trying to avoid

hypoglycaemia. No differences occurred in the frequency of severe hypoglycaemic

events resulting in seizure or coma, or in the frequency of severe hypoglycaemic events

resulting in admission to hospital. The lower the biochemical cutoff for hypoglycaemia

was set, the larger the reduction in hypoglycaemia with CGM, which might suggest that

patients in our trial took action only when their glucose concentration was already

3·9 mmol/L or lower, or perhaps they defined a lower threshold for self-treating

hypoglycaemia. Importantly, CGM did reduce the frequency of hypoglycaemic episodes

of less than 2·8 mmol/L, which can cause cognitive dysfunction as a result of neurogly-

copenia.28 We were not able to show a clinically relevant difference in self- reported

hypoglycaemia awareness between CGM and SMBG after 16 weeks, possibly because

CGM did not prevent all hypoglycaemia, but only reduced its duration and depth.

More rigorous avoidance of hypoglycaemic events for a longer period of time might be

needed to improve hypoglycaemia awareness.17 Our results suggest that CGM enables

patients to worry less about hypoglycaemia, but does not have a profound measurable

effect on other markers of quality of life.17

Ch

apte

r 4

100

These data add to the discussion about the value of CGM in preventing hypoglycaemia

in patients with impaired awareness of hypoglycaemia. Findings from a Cochrane

collaboration systematic review meta-analysis29 showed no difference in the incidence

rates of severe hypoglycaemia between CGM and SMBG (risk ratio 1·05, 95% CI

0·63–1·77); however, this analysis was based on data published up to June, 2011, and

included studies from which patients with impaired awareness of hypoglycaemia

had mostly been excluded. In a later observational study 1 year later, Choudhary and

colleagues14 reported a reduction of severe hypoglycaemia with CGM in patients with

impaired awareness of hypoglycaemia, reinforcing the need for targeted randomised

studies in patients with impaired awareness of hypoglycaemia. A trial using CGM

with low-glucose suspend seemed to support the idea of a benefit with CGM by

showing a reduction of severe hypoglycaemia in patients with impaired awareness

of hypoglycaemia.15 However, the population studied was quite young (mean age 18·6

years) and the reduction of severe hypoglycaemia was not significant when two outliers

in the youngest age groups were excluded from the analyses. Since patients with

impaired awareness of hypoglycaemia are usually older than 40 years and have had

diabetes for more than 25 years,16 these findings left important questions unanswered.

This issue was addressed in another randomised controlled trial (HypoCOMPaSS),

which specifically focused on adults with type 1 diabetes and impaired awareness

of hypo glycaemia.17 The investigators reported improved hypoglycaemia awareness

and glycaemic control from baseline to endpoint (24 weeks) with the use of extensive

patient guidance that included weekly contact, monthly follow-up visits, and use of a

bolus calculator to determine the insulin dose, whether or not an insulin pump was

used. However, no added benefit of CGM was shown. Importantly, sensors were used

for a median of 57% of the time in the HypoCOMPaSS study; only 17 of 42 individuals

achieved 80% sensor usage threshold, which is often considered the frequency required

for meaningful benefit. Our data add to these findings by showing that in typical adult

patients with long-standing type 1 diabetes and impaired awareness of hypoglycaemia,

CGM with median sensor usage of 89·4% (IQR 80·8–95·5) reduces severe hypoglycaemia.

When treating patients with type 1 diabetes who have impaired awareness

of hypoglycaemia and severe hypoglycaemia in clinical practice, health-care

professionals frequently first try to improve glycaemia by optimising self-management

(eg, by giving structured education about flexible insulin therapy) and changing the

insulin delivery method from MDI to CSII, before considering CGM, since structured

education programmes (such as DAFNE and BGAT) and CSII30-32 have been shown to

prevent severe hypoglycaemia and cost less than CGM. In our study population, 18

(35%) of 52 patients used carbohydrate counting and 23 (44%) of 52 patients were

on CSII. Our data showed equal benefit from CGM in both patients on CSII and MDI,

101

and in patients who did and did not use carbohydrate counting, with no interaction

for insulin treatment modality or the use of carbohydrate counting on the primary

outcome. This findings suggests that CGM can be used in various patients, including

those not willing or able to change to CSII or practise carbohydrate counting. Our study

has several strengths. Its crossover design removed between-patient variation, and the

washout period prevented any substantial carryover effects. Moreover, all data were

analysed on an intention-to-treat basis. Furthermore, we showed benefit of CGM in a

typical adult type 1 diabetes population with impaired awareness of hypoglycaemia,

with a mean age of 48·6 years (SD 11·6), median diabetes duration of 30·5 years (IQR

18·5 to 40·8), and a mean baseline HbA1c value of 7·5% (SD 0·8), which is similar to

adult patients with type 1 diabetes who have impaired awareness of hypoglycaemia

included in other trials.14,16,17 The treatment goals and guidance in our study were equal

in both intervention periods, with an equal number of follow-up visits and telephone

consultations during both periods.

A limitation of our study is that the CGM devices used in the trial might have been

outdated, since next-generation CGM systems came to market during the trial, with

improvements in lag time and accuracy, and with new features (eg, predicted low-glucose

suspension). Additionally, the masked and real-time CGM devices used in our trial are

known to differ somewhat in accuracy, which needs to be taken into account when

interpreting the CGM-derived data. The real-time CGM device is calibrated in real time,

but the masked CGM device is retrospectively calibrated (which allows the calibration

algorithm to use information both before and after the timepoint of interest to obtain an

optimum calibration to each reference point, leading to better accuracy). By contrast,

real-time CGM displays a glucose value in real-time and the calibration algorithm can

only use previous data for calibration. This difference might explain why the real-time

CGM device tends to report glucose concentrations that are lower than the reference

over the entire range of glucose values.20 However, if anything, this result would have

caused an overestimation of the reported CGM-derived hypoglycaemia during the

real-time CGM phase compared with the SMBG phase. The difference between CGM

and SMBG might therefore be larger than actually shown in this trial. Other limitations

were that the study could not be powered for severe hypoglycaemia as a primary

outcome, and that data for the frequency of adjustments to SMBG or therapy during the

intervention periods were not collected. CGM with predictive low-glucose suspension

could further reduce the incidence of severe hypoglycaemia in adult patients with

impaired awareness of hypoglycaemia, and clinical trials investigating this possibility

should be prioritised.

Ch

apte

r 4

102

In conclusion, in patients with type 1 diabetes and impaired awareness of

hypoglycaemia, CGM improved glycaemic control by decreasing both time spent

in a hypoglycaemic state and time spent in a hyperglycaemic state. Additionally,

it diminished severe hypoglycaemia. These results support the use of CGM in this

high-risk population.

103





J Med 2005; 353: 2643-53.

2. Orchard TJ, Nathan DM, Zinman B,

et al. Association between 7 years

of intensive treatment of type 1

diabetes and long-term mortality.

JAMA 2015; 313: 45-53.





registry. Diabetes Care 2015; 38:

971-8.



management of Type I and Type

II diabetes. Diabetologia 2002; 45:

937-48.




2008; 24: 87-92.




2014; 10: 711-22.








young adults. Diabetes Care 2008; 31:

922-6.





Diabetologia 2007; 50: 1140-7.

9. Cryer PE. Diverse causes of

hypoglycemia-associated autonomic

failure in diabetes. N Engl J Med

2004; 350: 2272-9.





Med 2008; 25: 501-4.






17: 697-703.













REFERENCES

Ch

apte

r 4

104

369: 224-32.







36: 4160-2.








clinical trial. JAMA 2013; 310: 1240-7.

16. Choudhary P, Geddes J, Freeman

JV, Emery CJ, Heller SR, Frier

BM. Frequency of biochemical

hypoglycaemia in adults with Type 1

diabetes with and without impaired

awareness of hypoglycaemia:

no identifiable differences using

continuous glucose monitoring.

Diabet Med 2010; 27: 666-72.











37: 2114-22.









Endocr Disord 2015; 15: 42.

19. American Diabetes Association.

Diagnosis and classification of


2014; 37 Suppl 1: S81-S90.

20. Calhoun P, Lum J, Beck RW, Kollman

C. Performance comparison of the

medtronic sof-sensor and enlite

glucose sensors in inpatient studies

of individuals with type 1 diabetes.


758-61.






Diabetes Obes Metab 2015; 17: 343-9.

22. ADA Guidelines 2015. Standards

of medical care in diabetes--2015.


S1-S93.







Care 1995; 18: 517-22.

24. Rodbard D. Interpretation of

105

continuous glucose monitoring data:

glycemic variability and quality of

glycemic control. Diabetes Technol

Ther 2009; 11 Suppl 1: S55-S67.



measure it. Diabetes 2013; 62: 1405-8.






795-800.






Diabetologia 2012; 55: 3155-62.




7: 493-503.






CD008101.







38: 1592-609.

31. Cox DJ, Gonder-Frederick L, Polonsky

W, Schlundt D, Kovatchev B, Clarke

W. Blood glucose awareness training

(BGAT-2): long-term benefits.

Diabetes Care 2001; 24: 637-42.

32. Shearer A, Bagust A, Sanderson D,

Heller S, Roberts S. Cost-effectiveness

of flexible intensive insulin

management to enable dietary

freedom in people with Type 1

diabetes in the UK. Diabet Med 2004;

21: 460-7.

Diabetes Technology and Therapeutics 2017 Oct;19(10):595-599

Cornelis A J van Beers, Maartje de Wit, Susanne J Kleijer,

Petronella H L M Geelhoed-Duijvestijn, J Hans DeVries, Mark H H Kramer

Michaela Diamant, Erik H Serné, Frank J Snoek

5

Continuous glucose

monitoring in patients with

type 1 diabetes and impaired


also effective in patients with

psychological distress?

Ch

apte

r 5

108

ABSTRACT

Background

To evaluate whether psychological distress modifies the effect of continuous glucose

monitoring (CGM) in patients with type 1 diabetes (T1D) and impaired awareness of

hypoglycaemia.

Methods

52 Patients with T1D and impaired awareness of hypoglycaemia participated in an

earlier reported randomized, crossover trial with two 16-week intervention periods

comparing CGM with self-monitoring of blood glucose (SMBG). During the CGM phase,

time spent in normoglycaemia (4-10 mmol/L), the primary outcome, was 9.6% higher

compared with the SMBG phase (p<0.0001). Psychological distress was operationalized

as either low emotional well-being (WHO-5 <50), high diabetes-related distress (PAID-5

≥8), and/or high fear of hypoglycaemia (HFS Worry > mean HFS Worry score + 1SD).

Modifying effects were assessed by analysing ‘psychological distress score × interven-

tion’-interaction effects.

Results

The low emotional well-being group and normal emotional well-being group showed

equal glycaemic outcomes during the CGM phase. High diabetes distress and elevated

fear of hypoglycaemia did not result in significant interaction effects for the glycaemic

outcomes.

Conclusions

CGM is equally effective in terms of glycaemic improvements in high versus low

distressed patients with T1D and impaired awareness of hypoglycaemia.

109

INTRODUCTION

Continuous glucose monitoring (CGM) can lower HbA1c1 and reduce the risk of

severe hypoglycaemia in patients with type 1 diabetes (T1D) and impaired awareness

of hypoglycemia.2-4 Impaired awareness of hypoglycaemia affects approximately 25%

of adults with T1D and increases the risk of severe hypoglycaemia substantially.5,6 CGM

may also help to reduce the adverse psychosocial effects of unpredictable recurrent

hypoglycaemia in T1D, but evidence is limited and inconsistent.7-9 In our recently

reported cross-over trial, CGM reduced the occurrence of severe hypoglycaemia by 59%

and significantly lowered worry of hypoglycemia.4 It is important to note that CGM can

aid patients to self-manage their diabetes more precisely, provided they are capable

of handling the device and data feedback adequately, with support of a diabetes

health care team.10,11 In this context, the question comes up whether CGM is suitable

and beneficial for patients with an unfavourable psychological profile, e.g. with

high psychological distress. Psychological distress is common in diabetes, including

low emotional well-being, high diabetes-related distress, and fear of hypoglycaemia,

negatively affecting patients’ daily self-care and glycaemic control.12-16 With CGM,

patients are faced with real-time feedback on blood glucose variation and alarms that

may be experienced as stressful and difficult to handle for those with pre-existing high

levels of distress.10,17,18 To the best of our knowledge, no trials investigating the effect of

CGM have studied the modifying role of psychological distress on treatment outcomes.

We therefore aimed to investigate whether psychological distress, operationalized

as either low emotional well-being, high diabetes-related distress, and/or elevated fear

of hypoglycaemia modify the effect of CGM on glycaemic outcomes in patients with

T1D and impaired awareness of hypoglycaemia.

Ch

apte

r 5

110

PATIENTS AND METHODS

The trial was a two-centre, randomized, open-label, crossover trial with a 12-week

wash-out period in between 16-week intervention periods, comparing CGM with

self-monitoring of blood glucose (SMBG).19 Fifty-two adults with T1D and impaired

awareness of hypoglycaemia as confirmed by a Gold score6 ≥4 and treated with either

continuous subcutaneous insulin infusion (CSII) or multiple daily injections (MDI)

were randomized to the CGM/SMBG (n=26) or the SMBG/CGM (n=26) sequence. During

the CGM phase, patients were instructed to use the CGM device continuously. During

the SMBG phase, patients were equipped with a masked CGM system which does not

display real-time glucose values. During both intervention periods, patients attended

monthly follow-up visits to review glucose data, based on CGM data during the CGM

phase and SMBG data during the SMBG phase, and make therapy changes accordingly.

These visits were followed by telephone consultations two weeks after each follow-up

visit comprising inquiry after adverse events, episodes of (severe) hypoglycaemia,

study device use and related technical issues, and check on medication changes.

Measurements

The primary endpoint of the trial was the difference in percentage of time spent in

normoglycaemia (4 – 10 mmol/L) between the CGM and the SMBG phase. Secondary

outcomes included the percentage of time patients spent in hypoglycaemia (≤3.9

mmol/L), average daily area under the curve ≤3.9 mmol/L (expressed as mmol/L×min),

frequency (episodes per week) of CGM-derived hypoglycaemic episodes ≤3.9 mmol/L

and <2.8 mmol/L, duration (min per episode) of CGM-derived hypoglycaemic episodes

≤3.9 mmol/L, severe hypoglycaemia (requiring third party assistance) and psychological

distress scores (WHO-5, PAID-5, HFS Worry). The primary and secondary outcomes of

the trial have been reported previously.4

At baseline and endpoint of both intervention periods, the psychological status of

the patients was assessed, covering three indicators of emotional distress: emotional

well-being, diabetes-distress, and fear of hypoglycaemia. Emotional well-being was

assessed using the World Health Organization Well-being Index 5 items questionnaire

(WHO-5).20 A score of <50 indicates clinically relevant low emotional well-being.20

Diabetes-related distress was assessed using the Problem Areas in Diabetes (PAID-5)

Short Form.21 A score of ≥8 indicates high diabetes distress. Fear of hypoglycaemia was

assessed with the Hypoglycaemia Fear Survey (HFS).22 The HFS comprises a Worry and

a Behaviour subscale. In this study we only used data from the Worry subscale, with

higher scores indicating more hypoglycaemia fear. A definite cut-off score for clinically

111

meaningful fear of hypoglycaemia in patients with T1D has yet to be established.

Therefore, in line with Hajos et al.23 we used the mean baseline HFS Worry score (38.4)

+ 1SD (19.2) as a cut-off value for elevated fear of hypoglycaemia.

Statistical methods

Using intention-to-treat analyses, we analysed the effect of CGM on glycaemic

outcomes using linear mixed-model analysis with the glycaemic outcome as the

dependent variable, the treatment group (CGM or SMBG) as a fixed factor, and the

participant as a random factor. In order to demonstrate whether indicators of

psychological distress modified the effect of CGM on glycaemic outcomes, we included

a two-way interaction term (psychological distress score × intervention [CGM or

SMBG]) in the linear mixed model. If a significant interaction effect was seen, we

conducted post-hoc analyses on the subgroups, using similar linear mixed models.

Statistical significance for interaction effects was set at 0.1 (2-tailed) and statistical

significance for all other analyses was set at 0.05 (2-tailed). Patients were scored as

having low emotional well-being if the WHO-5 score was <50 at baseline of the CGM

phase or at baseline of the SMBG phase. Likewise, patients were scored as having high

diabetes distress if the PAID-5 score was ≥8 at baseline of the CGM phase or at baseline

of the SMBG phase. Elevated fear of hypoglycaemia was reported in patients with a

HFS Worry score >57.6 at baseline of the CGM phase or at baseline of the SMBG phase.

All analyses were conducted using SPSS 22.0 for windows (IBM SPSS Inc., Chicago, IL,

USA).

Ch

apte

r 5

112

RESULTS

The mean age of the population was 48.6 ± 11.6 years, 24 (46.2%) were female,

median duration of diabetes was 30.5 years (IQR 18.5 to 40.8), 23 patients (44.2%) were

treated with CSII, mean HbA1c was 7.5% ± 0.8% and all patients reported impaired

awareness of hypoglycaemia, with a mean Gold score of 5.4 ± 0.7. The intention-to-treat

population consisted of all 52 randomized patients.

Questionnaire completion was high at baseline and endpoint of both intervention

periods (baseline CGM 50 patients [96%]; endpoint CGM 48 patients [92%]; baseline

SMBG 49 patients [94%]; endpoint SMBG 47 patients [90%]).Thirteen patients (25%)

reported low emotional well-being (WHO-5 score <50) indicative of depressive affect

at baseline of the CGM phase or at baseline from the SMBG phase, 19 (36.5%) patients

reported high diabetes-distress (PAID-5 score ≥8) at baseline of the CGM phase or

at baseline from the SMBG phase, and 13 patients (25%) reported elevated fear of

hypoglycaemia (HFS Worry score > 57.6) at baseline of the CGM phase or at baseline

from the SMBG phase. Mean PAID-5 score was 6.0 (95% CI 5.0 to 7.1) at baseline of the

CGM phase and 5.9 (95% CI 4.9 to 7.0) at baseline of the SMBG phase. Mean WHO-5

score was 65.9 (95% CI 60.6 to 71.2) at baseline of the CGM phase and 62.9 (95% CI

57.6 to 68.3) at baseline of the SMBG phase. Mean HFS Worry score was 38.9 (95% CI

33.5 to 44.2) at baseline of the CGM phase and 38.2 (95% CI 32.8 to 43.6) at baseline of

the SMBG phase. The 16-week HFS Worry score was significantly lower after the CGM

phase compared with the 16-week HFS Worry score after the SMBG phase (32.5 vs

38.9; Δ 6.4 [95% CI 1.4 to 11.4], p=0.014), but no significant differences were seen in the

16-week WHO-5 or PAID-5 scores comparing CGM to SMBG (data not shown).

The interaction term ‘low emotional well-being × intervention’ demonstrated

significant effects for the primary outcome of the trial (time spent in normoglycaemia,

p=0.024), and for the hypoglycaemia related outcomes (time spent in hypoglycaemia,

p=0.009; frequency of hypoglycaemic events <3.9 mmol/L, p=0.098; frequency of

hypoglycaemic events <2.8 mmol/L, p=0.019; duration of hypoglycaemia, p=0.079; and

area under the curve in hypoglycaemia, p=0.007. Table 1 shows post-hoc subgroup

analyses of the outcomes with significant interaction effects (emotional well-being

× intervention). Although still significant for the primary outcome (time spent in

normoglycaemia), the effect sizes of CGM versus SMBG on glycaemic outcomes was

consistently lower in the 13 patients with low emotional well-being compared with

the 39 patients with normal emotional well-being. These lower effect sizes were

mainly caused by a higher level of glycaemic control during the SMBG phase in the low

emotional well-being group, rather than a lower level of glycaemic control during the

113

Lo

w e

mo

tio

na

l w

ell-

bei

ng

(WH

O-5

<50

, n=1

3)

Gly

caem

ic o

utc

om

eC

GM

SM

BG

Mea

n d

iffe

ren

ce

(95%

CI)

*P

va

lues

Per

cen

t of

tim

e sp

ent

wit

h g

luco

se le

vel i

n r

ange

4.0-

10 m

mol

/L65

.0 (

60.2

to

69.8

)58

.4 (

53.6

to

63.2

)6.

6 (2

.5 t

o 10

.7)

0.00

4

≤3.9

mm

ol/L

6.8

(3.7

to

9.9)

9.0

(5.9

to

12.1

)2.

2 (-

0.4

to 4

.9)

0.09

1

Fre

qu

ency

of

hyp

ogly

caem

ia ≤

3.9

mm

ol/L

(ev

ents

/wee

k)9.

2 (6

.7 t

o 11

.7)

9.2

(6.7

to

11.7

)0.

1 (-

2.2

to 2

.1)

0.93

9

Fre

qu

ency

of

hyp

ogly

caem

ia ≤

2.8

mm

ol/L

(ev

ents

/wee

k)2.

8 (1

.2 t

o 4.

5)3.

5 (1

.8 t

o 5.

2)0.

6 (-

0.9

to 2

.1)

0.38

0

Du

rati

on o

f C

GM

-der

ived

hyp

ogly

caem

ic e

ven

ts (

min

/eve

nt)

64.9

(51

.6 t

o 78

.2)

94.5

(81

.2 t

o 10

7.8)

29.6

(16

.2 t

o 43

.1)

0.00

1

AU

C ≤

3.9

mm

ol/L

per

24

hou

rs (

mm

ol/L

x m

inu

tes)

69.3

(30

.7 t

o 10

7.9)

59.8

(51

.2 t

o 12

8.5)

20.5

(-1

2.8

to 5

3.8)

0.20

0

No

rma

l em

oti

on

al

wel

l-b

ein

g (W

HO

-5 ≥

50, n

=39)

Gly

caem

ic o

utc

om

eC

GM

SM

BG

Mea

n d

iffe

ren

ce

(95%

CI)

*P

va

lues

Per

cen

t of

tim

e sp

ent

wit

h g

luco

se le

vel i

n r

ange

4.0-

10 m

mol

/L64

.7 (

62.1

to

67.3

)53

.9 (

51.3

to

56.5

)10

.8 (

9.1

to 1

2.5)

<0.0

001

≤3.9

mm

ol/L

6.6

(4.7

to

8.5)

12.4

(10

.5 t

o 14

.3)

5.8

(4.5

to

7.1)

<0.0

001

Fre

qu

ency

of

hyp

ogly

caem

ia ≤

3.9

mm

ol/L

(ev

ents

/wee

k)10

.2 (

8.5

to 1

1.9)

12.0

(10

.3 t

o 13

.7)

1.7

(0.7

to

2.7)

0.00

1

Fre

qu

ency

of

hyp

ogly

caem

ia ≤

2.8

mm

ol/L

(ev

ents

/wee

k)1.

9 (1

.1 t

o 2.

7)4.

2 (3

.4 t

o 5.

0)2.

3 (1

.6 t

o 2.

9)<0

.000

1

Du

rati

on o

f C

GM

-der

ived

hyp

ogly

caem

ic e

ven

ts (

min

/eve

nt)

58.1

(51

.3 t

o 64

.8)

100.

5 (9

3.7

to 1

0.7.

2)42

.4 (

34.4

to

50.4

)<0

.000

1

AU

C ≤

3.9

mm

ol/L

per

24

hou

rs (

mm

ol/L

x m

inu

tes)

59.8

(38

.5 t

o 81

.3)

126.

6 (1

05.1

to

148.

0)66

.7 (

50.7

to

82.8

)<0

.000

1

Dat

a ar

e m

ean

(95

% C

I). *

Res

ult

of

lin

ear

mix

ed-m

odel

an

alys

is w

ith

th

e gl

ycae

mic

ou

tcom

e th

e d

epen

den

t va

riab

le, t

he

trea

tmen

t gr

oup

(C

GM

or

SMB

G)

as a

fix

ed f

acto

r, a

nd

th

e p

arti

cip

ant

as a

ran

dom

fac

tor.

AU

C: A

rea

un

der

th

e cu

rve;

CG

M: C

onti

nu

ous

glu

cose

mon

itor

ing;

SM

BG

: Sel

f-m

onit

orin

g of

blo

od g

luco

se.

Ta

ble

1. P

ost

-ho

c a

na

lyse

s o

f it

ems

wit

h s

ign

ific

an

t ‘e

mo

tio

na

l w

ell-

bei

ng

× in

terv

enti

on

’ in

tera

ctio

n e

ffec

ts

Ch

apte

r 5

114

CGM phase in the low-emotional well-being group (Table 1). The interaction term ‘high

diabetes distress × intervention’ resulted in one significant interaction effect for time

spent in normoglycaemia (p=0.05), but no significant interaction effects were seen for

the other glycaemic outcomes. Again, post-hoc analyses showed no relevant difference

in time spent in normoglycaemia during the CGM phase between patients with high

diabetes distress compared with those without high diabetes distress (65.4% [95% CI

61.8 – 69.1] vs. 64.5% [61.5 to 67.4]). No significant interaction effects were found for

the interaction term ‘elevated fear of hypoglycaemia × intervention’.

115

DISCUSSION

We report on the possible modifying effect of psychological distress, specifically

emotional well-being, diabetes-related distress and fear of hypoglycaemia on CGM

outcomes as recently reported in a trial in patients with T1D and impaired awareness

of hypoglycaemia comparing CGM versus SMBG.4 We observed lower effect sizes in

the low emotional well-being group compared with the ‘normal’ emotional well-being

group, which could suggest that CGM is less effective in patients with low emotional

well-being. However, the effect of CGM on the primary outcome (time spent in

normoglycaemia) remained significant also in patients with low emotional wellbeing.

Furthermore, the lower effect sizes were mainly due to better glycaemic control during

the SMBG phase, rather than to poorer glycaemic control during the CGM phase. No

relevant or consistent interaction effects were observed between diabetes-related

distress or fear of hypoglycaemia and the glycaemic outcomes.

With regard to the external validity, our study population was relatively small, well

educated, and lacked ethnic minorities. Otherwise the clinical profile of our study

sample does not suggest selection bias. The mean age, HbA1c, and diabetes duration

of our study population seem comparable to the baseline characteristics of patients

included in other trials in diabetes patients with impaired awareness of hypoglycaemia

or frequent severe hypoglycemia.2,3,5,24 The mean total WHO-5 score and mean total

PAID-5 score are comparable to the WHO-5 and PAID-5 scores reported in previously

published trials in type 1 and type 2 diabetes patients without established impaired

awareness of hypoglycemia.7,25,26 Not unexpectedly, the mean HFS Worry score

indicated that the patients in our study report to worry more about hypoglycaemia

relative to T1D patients without established impaired awareness of hypoglycemia.7,8,27

Further research in larger and more diverse T1D patient populations is warranted.

The cross-over design is a limitation, as the level of psychological distress can change

during the course of a study period, although baseline scores at the two respective

study periods were not dissimilar. Future studies should replicate our findings in

studies not using a crossover design. We did not find an indication that psychological

distress adversely affects CGM use in this patient population with IAH . However, any

current psychiatric disorder, including major depression, was an exclusion criterion

limiting generalizability. Further research should help clarify if and to what extent

GCM outcomes are negatively impacted when used in patients with more severe

psychological distress. Also, all patients had impaired awareness of hypoglycaemia,

limiting generalization to other patient populations. The level of support during the

trial was relatively high compared with usual care. It could be hypothesized that

Ch

apte

r 5

116

patients with psychological distress require a higher level of support compared with

those without psychological distress and that this higher level of support cannot be met

in usual clinical practice.

In conclusion, low emotional well-being, high diabetes-related distress and fear of

hypoglycaemia do not substantially modify the effect of CGM on glycaemic outcomes

in T1D patients with impaired awareness of hypoglycaemia. We can therefore expect

these more psychologically vulnerable patients to benefit from CGM as much as those

with a more favourable psychological profile.

117








patient data. BMJ 2011;343:d3805.





hypoglycemia-unaware patients

with type 1 diabetes. Diabetes Care

2013;36:4160-4162.

3. Ly TT, Nicholas JA, Retterath A, et al.

Effect of sensor-augmented insulin

pump therapy and automated insulin




clinical trial. JAMA 2013;310:1240-

1247.

4. van Beers CA, DeVries JH, Kleijer

SJ, et al. Continuous glucose

monitoring for patients with type 1

diabetes and impaired awareness

of hypoglycaemia (IN CONTROL): a

randomised, open-label, crossover

trial. Lancet Diabetes Endocrinol

2016.

5. Geddes J, Schopman JE, Zammitt

NN, et al. Prevalence of impaired



Med 2008;25:501-504.





hypoglycemia. Diabetes Care

1994;17:697-703.






controlled trial. Diabet Med

2011;28:1158-1167.







Diabetes Care 2010;33:2175-2177.

9. Polonsky WH, Hessler D, Ruedy

KJ, et al. The Impact of Continuous

Glucose Monitoring on Markers of

Quality of Life in Adults With Type

1 Diabetes: Further Findings From

the DIAMOND Randomized Clinical

Trial. Diabetes Care 2017.





2010;27:1060-1065.11. Peters AL, Ahmann AJ, Battelino

T, et al. Diabetes Technology-

REFERENCES

Ch

apte

r 5

118

Continuous Subcutaneous Insulin

Infusion Therapy and Continuous

Glucose Monitoring in Adults: An

Endocrine Society Clinical Practice

Guideline. J Clin Endocrinol Metab

2016;101:3922-3937.



with diabetes. J Clin Psychol

2001;57:457-478.


KE, et al. Depression and poor

glycemic control: a meta-analytic

review of the literature. Diabetes

Care 2000;23:934-942.

14. Wild D, von MR, Brohan E, et al.

A critical review of the literature

on fear of hypoglycemia in



Patient Educ Couns 2007;68:10-15.

15. Barendse S, Singh H, Frier BM, et

al. The impact of hypoglycaemia on

quality of life and related patient-

reported outcomes in Type 2

diabetes: a narrative review. Diabet

Med 2012;29:293-302.

16. Fisher L, Hessler D, Polonsky W, et al.




Complications 2016;30:1123-1128.


KM, et al. Diabetes Device Use

in Adults With Type 1 Diabetes:

Barriers to Uptake and Potential

Intervention Targets. Diabetes Care

2017;40:181-187.






2015;38:544-550.









Endocr Disord 2015;15:42.

20. Hajos TR, Pouwer F, Skovlund SE,

et al. Psychometric and screening

properties of the WHO-5 well-being

index in adult outpatients with Type

1 or Type 2 diabetes mellitus. Diabet

Med 2013;30:e63-e69.

21. McGuire BE, Morrison TG, Hermanns

N, et al. Short-form measures of

diabetes-related emotional distress:

the Problem Areas in Diabetes Scale

(PAID)-5 and PAID-1. Diabetologia

2010;53:66-69.

22. Cox DJ, Irvine A, Gonder-Frederick

L, et al. Fear of hypoglycemia:

quantification, validation,

and utilization. Diabetes Care

1987;10:617-621.

23. Hajos TR, Polonsky WH, Pouwer F,

et al. Toward defining a cutoff score

for elevated fear of hypoglycemia

on the hypoglycemia fear survey

119

worry subscale in patients with

type 2 diabetes. Diabetes Care

2014;37:102-108.

24. Rondags SM, de Wit M, Twisk JW,

et al. Effectiveness of HypoAware,

a Brief Partly Web-Based

Psychoeducational Intervention

for Adults With Type 1 and

Insulin-Treated Type 2 Diabetes

and Problematic Hypoglycemia: A

Cluster Randomized Controlled Trial.

Diabetes Care 2016.

25. Snoek FJ, Kersch NY, Eldrup E, et

al. Monitoring of Individual Needs

in Diabetes (MIND): baseline data

from the Cross-National Diabetes

Attitudes, Wishes, and Needs

(DAWN) MIND study. Diabetes Care

2011;34:601-603.

26. Nicolucci A, Kovacs BK, Holt RI,

et al. Diabetes Attitudes, Wishes

and Needs second study (DAWN2):

cross-national benchmarking of

diabetes-related psychosocial

outcomes for people with diabetes.

Diabet Med 2013;30:767-777.

27. Rubin RR, Peyrot M. Health-related

quality of life and treatment

satisfaction in the Sensor-Augmented

Pump Therapy for A1C Reduction 3

(STAR 3) trial. Diabetes Technol Ther

2012;14:143-151.

Diabetic Medicine 2017 Oct;34(10):1470-1476

Cornelis A J van Beers*, Anne F Vloemans*, Maartje de Wit, Wilmy Cleijne

Stefanie M Rondags, Petronella H L M Geelhoed-Duijvestijn,

Mark H H Kramer, Erik H Serné, Frank J Snoek

*joint first authors

6

Keeping safe.

Continuous glucose

monitoring in persons with

type 1 diabetes and impaired


a qualitative study

Ch

apte

r 6

122

ABSTRACT

Aims

To further our understanding of individual use and experience of continuous

glucose monitoring (CGM) in adults with type 1 diabetes and impaired awareness of

hypoglycaemia, we conducted a qualitative study supplementary to a randomised

controlled trial, using semi-structured interviews.

Methods

Twenty-three participants of the IN CONTROL trial were interviewed within four

weeks after the last study visit. The interview centred around experiences of CGM,

taking into account the person’s expectations prior to the trial. The interview was

semi-structured, using open-ended questions and if needed, prompts were offered to

elicit further responses. Using thematic analysis, the interview transcripts were coded

independently by three members of the research team. The consolidated criteria for

reporting qualitative research (COREQ) were followed.

Results

CGM was overall experienced as helpful in gaining more insight into glucose

variability and temporarily improved sense of control, reduced distress and made

participants less dependent on others. However, some participants experienced

confrontation with CGM output as intrusive, whilst some reported frustration due to

failing technique and difficulty trusting the device. Participants reported active and

passive self-management behaviours mirroring individual differences in attitudes

and coping styles.

Conclusions

In adults with type 1 diabetes at risk for recurrent hypoglycaemia due to impaired

awareness of hypoglycaemia, CGM use enhances a sense of control and safety for

most but not all. Future studies should further explore differential use of CGM in this

population in the context of active and passive self-management styles.

123

INTRODUCTION

Hypoglycaemia is a well-recognised barrier to achieving glycaemic targets in

type 1 diabetes (T1DM), adversely affecting peoples’ well-being.1,2 In order to avert

hypoglycaemia, people with T1DM require frequent feedback on their blood glucose

fluctuations allowing for preventive measures and timely corrections. To this end,

real-time continuous glucose monitoring (CGM) has clear advantages over conventional

self-monitoring of blood glucose (SMBG) as it displays semi-continuous interstitial

glucose values, visualized glucose trends and the opportunity to set alarms to warn

for impending hypoglycaemia and hyperglycaemia. Indeed, studies have consistently

shown that CGM systems lower HbA1c values3 and reduce severe hypoglycaemia

(hypoglycaemia requiring third party assistance) in people with T1DM,4-6 associated

with less of worrying about hypoglycaemia.6,7

As noted by Kubiak et al.8 CGM systems in itself do not directly impact glucose

regulation (unless combined with a pump with a low-glucose suspend function5),

but can prompt timely self-management responses that can help gain control over

undesirable blood glucose fluctuations. A large-scale survey in CGM users in the US

revealed that over 80% of the respondents believed that CGM facilitated their self-ma-

nagement with benefits including hypoglycaemia avoidance and safety.9 On the

other hand, CGM and its alarms can be experienced as intrusive and patients can feel

overwhelmed by the precise feedback they receive on blood glucose fluctuations. In

this context, research has identified several themes related to CGM, such as ‘coping

with frustrations’ (confrontation with unexpected glucose values; false alarms), ‘use of

CGM information’ (when to act and how), and ‘technical hassles’ (failures of CGM).10,11

To date, in persons with intact hypoglycaemia awareness10-12 CGM has shown mixed

results with regard to quality of life outcomes.6,7,13 In persons with T1DM at risk of

recurrent severe hypoglycaemia, the perceived benefits of CGM are likely to be

determined by the actual experience of protection from hypoglycaemic episodes,

translating into feeling more safe and less need to worry about losing control.

We previously reported on the main outcomes of the IN CONTROL cross-over trial,6

showing that CGM compared to SMBG results in more time spent in normoglycaemia

and reduced severe hypoglycaemia in persons with T1DM and impaired awareness

of hypoglycaemia. As to patient-reported outcomes, those treated with CGM reported

significantly less worry about hypoglycaemia relative to SMBG, but no other differences

in markers of quality of life were observed.

Ch

apte

r 6

124

To further our understanding of the use, perceptions and experiences of CGM in

this typical population at high risk of hypoglycaemia, we conducted a supplementary

qualitative study using semi-structured interviews. The study was aimed to explore

common themes as well as individual differences with respect to expectations and

perceived benefits of CGM in this selected patient group.

125

METHODS

Sample and procedures

IN CONTROL was a randomised, crossover trial with a 12-week wash-out in between

two 16-week intervention periods investigating the effect of CGM on glycaemic

outcomes and quality of life in 52 adults with T1DM and impaired awareness of

hypoglycaemia. Mean age was 48.6 years (SD 11.6); diabetes duration 30.5 years (IQR

18.5 – 40.8); Gold score 5.4 (SD 0.7) (indicating impaired awareness of hypoglycaemia).

During the intervention period the participants used a real-time CGM device (CGM

phase) or standard self-monitoring of blood glucose (SMBG) during the control period.

Twenty-six participants were randomised to the CGM/SMBG sequence and twenty-six

participants to the SMBG/CGM sequence. After approval of the ethical committee of

the VU University Medical Center (VUMC), twenty-three participants were invited for

an interview with a spread of gender, age and insulin treatment modality (injections,

pump). All agreed and gave written consent. Of these 23 participants, 15 were

randomised to the SMBG/CGM sequence and 8 were randomised to the CGM/SMBG

sequence. At the last study visit (after 16 weeks of CGM use) participants had to return

the CGM device in accordance with the protocol. Participants were interviewed within

four weeks after the last study visit.

Interview

The interview schedule was developed by the research team, informed by previous

work in the field.10-12 Interview questions were open-ended, pertaining to expectations

and experiences of the participant using CGM (either pre- or post SMBG phase).

If open-ended questions elicited little response, prompts were offered (Table 1). All

participants were individually interviewed by one researcher (SMR, psychologist),

between April 24th, 2015 and March 21th, 2016. Interviews lasted 20-50 minutes and

were audio-recorded and transcribed verbatim anonymously.

Data analysis

We took an exploratory approach to the interview data using thematic analysis.14

The interview transcripts were coded independently by three members of the research

team. The coders independently selected sections from the interview relevant to the

research objectives and coded these into key words and themes. After every three

interviews a team meeting was planned to discuss the thematic analysis of responses

and deviant cases, compare interpretations, and reach agreement on recurring themes.

Confirmability was enhanced by document procedures for checking and rechecking of

the data. Credibility of the data interpretation was ensured by investor triangulation,

a process of analysing data with multiple perspectives.15 The final theoretical

Ch

apte

r 6

126

framework reflected the topics explored with participants and emergent themes. This

framework was developed once all data had independently and jointly been reviewed

and consensus had been reached. Transcripts of participants randomised to the SMBG/

CGM sequence and the CGM/SMBG sequence were separately coded to explore possible

differences in emerging themes. To ensure transferability of the data, research

context and assumptions were described.15 Finally, one researcher re-examined the

raw transcripts and data analyse procedures to confirm that all relevant data were

indeed reflected in the coding and to ensure dependability. The consolidated criteria

for reporting qualitative research (COREQ) were followed.16

Interview

You have participated in the IN CONTROL study, a study aimed to test the effect of CGM and SMBG on reduction of

hypoglycaemia.

1. Was that your main reason for joining the study? Or were there other reasons? If so, can you describe them? (e.g.

curiosity, aiming for better control0

2. What would you consider the most burdensome: hypo- or hyperglycaemia. Can you elaborate?

• What do you experience when having a hypo?

• What do you experience when having a hyper?

3. What were your expectations regarding the real-time sensor/the study?

• Better diabetes regulation

• Improvement of well-being

• Regarding social environment?

4. To what extent have your expectations been met?

• Did you trust the real-time sensor

• Accuracy

• Alarms

• Lag-time

• Do you have the feeling that the real-time sensor gives you more control?

• Do you have the feeling that you were more occupied with your diabetes, due to the real-time sensor?

What was your experience?

• How user-friendly was the system?

• With the sensor you received a lot of information, trends and graphics. How was that for you? Did you feel

supported enough? Do you have suggestions for improvement?

5. I have heard what you said. Summary – and can you tell perhaps a bit more about the positives and negatives of a

real-time sensor from your perspective?

6. If you were to decide, what would you change in the system, if anything?

7. You had to hand in the sensor at some point in time, how was that for you?

8. Suppose the sensor would be reimbursed for patients with less awareness of hypo symptoms, would you recommend

it to others?

9. Have I forgotten to ask anything that you would consider important?

Table 1. Interview

127

RESULTS

Participants

Baseline characteristics of the 23 participants are represented in Table 2 and

comparable to the total patient population of the IN CONTROL trial.6 No considerable

differences were observed in themes that emerged from participants randomised to the

SMBG/CGM sequence versus those randomised to the CGM/SMBG sequence, from men

versus women, or from participants younger than the median age (48.5 years) versus

those older than the median age. All patients were new to CGM, with the exception of

one participant who had once used CGM during a marathon more than six months

before randomisation.

Population (n=23)

Female 10 (43.5%)

Age (years) 47.7 (13.0)

BMI (kg/m2) 24.6 (3.6)

Diabetes duration (years) 28.1 (11.2, 41.7)

HbA1c (mmol/mol) 59 (8)

HbA1c (%) 7.1 (0.7)

Insulin treatment modality, CSII 10 (43.5%)

Duration of CSII therapy (years) 13.3 (4.9, 19.9)

Hypoglycaemia awareness score6 5.4 (0.8)

BMI: Body mass index; CSII: Continuous subcutaneous insulin infusion. Data is given as n (%), mean (SD) or median

(25%, 75%).

Table 2. Characteristics of people participating in interview

Motives and expectations

At the start of the interview people were asked to elaborate on their motives to

participate in the IN CONTROL trial. As expected, the main reason to join the study was

that they felt a need for protection, given their reduced ability to perceive hypoglycaemia

and consequently experiencing frequent devastating hypoglycaemic episodes. A few

participants mentioned to have been in a traffic accident due to hypoglycaemia.

[Male, age 32]: “I don’t have any sense of a hypo coming, it was just there instantly and

it was done. … I had an accident with a scooter[…] that was a sort of eye-opener to me”

Participants entered the study with the expectation that wearing the CGM device

would help them better understand and gain control over fluctuations in their blood

Ch

apte

r 6

128

glucose levels. They were expecting CGM to warn them against lowering blood glucose

levels, allowing them to feel more secure and less dependent on others. This in turn

was expected to also offer relief to their significant others for whom hypoglycaemia

also had become a major worry in their lives.

Experiences with CGM

Three main themes emerged from the interviews: ‘sense of control and well-being’,

‘ease of use’ and ‘handling of CGM information’.

Sense of control and well-being

The majority of participants experienced more control over their diabetes while

using CGM, in line with their expectations.

[Male, age 42]: “I can only be positive and in two sentences. I was in the middle of a

diabetes-burnout and I am now sure of myself and that is what it brought me. This study

I think is called ‘IN CONTROL’, well that’s it.”

Participants gained better insight into daily trends and patterns, and the impact of

exercise, meals, sleep, and stress on blood glucose values.

[Male, age 65]: “Especially seeing the values in context gave a sense of control. So is the

value going up or down, what were the previous values. That is most important”.

Several participants reported an increased sense of control as a result of the ability to

better predict hypoglycaemia with the assistance of CGM. As a result less (unexpected)

blood glucose fluctuations and profound (nocturnal) hypoglycaemia were experienced

and participants felt their diabetes was more ‘stable’ or ‘constant’. Knowing that they

would get an alert prior to blood glucose falling too low was reassuring for most persons.

The majority of participants found that CGM use had improved their well-being. It

was mentioned that CGM served as a ‘buddy’ or ‘guardian’ that made them feel less

distressed, more confident, and secure.

[Female, age 56]: “…that felt like I had a buddy who watched over me. That was the

feeling I had.”

Importantly, the protection offered by CGM helped patients to be less dependent on

spouses and family members; participants reported their significant others to be less

worried (e.g. regarding nocturnal hypoglycaemia) while using CGM and sleep better

at night, even with the alarms. Participants experienced their significant others to

be more supportive and understanding of the complexity of attaining good glucose

129

control.

For some, CGM negatively affected their well-being and sense of control, as a result

of being confronted with daily and between-day glucose fluctuations that they had

previously not been aware of.

[Male, age 37]: “Fluctuations you previously were not aware of , you do notice with the

sensor. And that is frustrating […] you keep thinking why does it rise”

Some participants were annoyed by frequent alarms, especially at night-time or

during working hours and found that CGM disturbed their normal daily life adding to

the psychosocial burden of living with T1DM. A few participants did not experience

an increased sense of control, mainly because they had difficulty trusting the device

after having experienced a technical failure. After discontinuation of CGM at the end

of the study period, most participants commented that they missed the sense of control

that CGM had provided them. The few participants who were less positive about CGM

felt comfortable or even relieved after returning the CGM device. Several participants

expressed the hope that the IN CONTROL trial would result in full reimbursement of

CGM for people with T1DM and impaired hypoglycaemia awareness.

Ease of use

The majority of participants were positive about the ease of use of CGM and its

features, although the design of the sensor was considered somewhat outdated. The

possibility of frequent, effortless checking of their blood glucose was appreciated.

However, some mentioned they felt annoyed when inaccurate CGM results proved to

be inaccurate. There were complaints about the need for calibration of the CGM system

with SMBG and the unpleasant or painful insertion of the sensor. Also, alarms were

perceived by some as too soon, too loud or too often and inconvenient during sleep,

work and/or exercise. Some physical concerns and inconveniences were mentioned,

such as wearing the sensor and with certain cloths (or without), during exercise, sleep

and while being intimate with their partner. Some commented on skin irritation and

problems attaching the sensor to the skin.

Handling of CGM information

Widely different attitudes and behavioural responses were noted with regard to

whether or not trends were used for treatment decisions, how participants responded

to alarms and whether or not the alarms were switched on during day and night.

Trusting the sensor seemed to play an important role in this context. Some mentioned

to feel annoyed or embarrassed when having to respond to the CGM in the presence of

Ch

apte

r 6

130

others. Roughly half of the participants appreciated the alarms and switched them on

during day- and night-time.

[Female, age 56]: “ I very much liked the alarms. I liked that I was being warned when

,my blood sugar was too high or too low […] It took out the extremes.”

Some switched off the alarms during certain activities (e.g. social events or work)

or while sleeping. Others considered hypoglycaemic alarms at 4.5 mmol/L as not (yet)

relevant for taking action.

[Female, age 31 ] : “Sometimes when busy during the day, working with clients, it was

annoying that at 4.5 it started to scream -- that is how I called it […] I didn’t do anything

with those 4.5 alarms, but also didn’t shut them off.”

CGM prompted some participants to be more actively engaged in self-managing

their diabetes (e.g. adjusting insulin dose), while a few participants explained that CGM

had the opposite effect: seeing their glucose values on a semi-continuous basis and by

setting alarms led them to become more passive.

[Male, age 61]: “And sometimes you would just rely on it, that is a disadvantage, that

you think it will beep. I would just wait till it beeps.”

It was mentioned by a few that they had become ‘somewhat obsessed’ by continuously

checking the CGM, which added to ‘feeling like a diabetic’ instead of ‘knowing you

have diabetes’. After discontinuation of CGM at the end of the study period, some

participants noted small improvements in their self-management behaviours, for

example performing finger pricks more frequently and adjusting insulin dose in

specific situations (e.g. sports):

[Male, age 61]: ‘…though, I do notice, it’s better to check more often, because the

numbers tell the tale…. In the past I checked once in the morning, once during the

afternoon and once before I went to bed. But nowadays I measure, well uh, at least six

times. […] And this, I believe, provides guidance in the future.’

131

DISCUSSION

The benefits experienced with CGM in patients with type 1 diabetes and impaired

hypoglycaemia awareness clearly relate to a better sense of control and reduced stress

around hypoglycaemia, positively impacting quality of life. Most participants found

CGM and its features helpful to achieve better glucose control, i.e. ‘being more stable’

and having less or no profound hypoglycaemia.

Most participants had realistic expectations of CGM when entering the study, which

is thought to be important for CGM adherence.8 Unrealistic expectations may lead to

disappointment and discontinuation of CGM.17 Indeed, during the IN CONTROL trial,

median sensor use was high (89.4%) and the CGM effect was sustained throughout the

intervention period.6 This was achieved in context of a clinical trial, where participants

were motivated, well informed and received personal education when needed in order

to optimise use of CGM.

Our findings indicate that the positive experiences clearly outweigh the negative,

that seem partly related to established technical issues that go hand in hand with todays

marketed CGM devices.10,11 Most frequently mentioned problem was the perceived

(in)accuracy of CGM, which in some participants resulted in annoyance and difficulty

trusting the CGM device. In addition, frequent and (perceived) false alarms were an

important source of frustration. These findings concord with previous psychosocial

research that highlighted ‘coping with frustrations’ and ‘technical hassles’ as relevant

themes.10,11

CGM triggered some patients to engage in understanding trends and more actively

manage their glucose levels, while others reported to respond more passively and

wait for the alarm to go off, before getting into action. Interestingly, these individual

differences in coping with CGM do not predict glycaemic outcomes per se, as all study

participants, but one, responded well to real-time CGM in terms of time spent in range

and reduced hypoglycaemia, at least for the study period.6 This questions Ritholz and

colleagues’ conclusion that people with T1DM with active self-management behaviours

and self-controlled coping styles use CGM features more effectively.10 Future studies

are needed to examine how attitudes and coping styles influence self-management

behaviour during as well as after using CGM and how these may enhance improvement

of glycaemic control, both in people with and without impaired awareness of

hypoglycaemia.

Ch

apte

r 6

132

Our findings are largely in line with previously published qualitative research

exploring participants’ experiences with CGM not necessarily affected by impaired

awareness of hypoglycaemi.10,11,18,19 Pickup and colleagues explored unprompted

patient narratives on CGM use at home of children and adults with T1DM and found

overwhelmingly positive experiences of CGM.11 Older adults at risk for hypoglycaemia

and hypoglycaemia unawareness, also reported improvement of well-being next

to feelings of safety while using CGM.19 Importantly, participants reported not only

the benefits of CGM for themselves but also their partners, who are also affected by

worries about (nocturnal) hypoglycaemia and the burden of having to treat the patient

in order to restore consciousness.

The improved sense of control many participants (and partners) experienced concurs

with the demonstrated objective glycaemic improvement such as more time spent in

the target range, reduced glycaemic variability, and less frequent nocturnal and severe

hypoglycaemia.5,6 Interestingly, according to the majority of participants, the improved

sense of control washed away after discontinuation of CGM, implicating there was no

psychological legacy effect. Indeed , the improved level of glycaemic control (time

in range) was lost after switching from CGM to SMGB.6 It would appear that for this

selected group of patients, continued use of CGM is advised, which obviously has

cost implications. It would seem worthwhile to test whether offering Blood Glucose

Awareness Training,20,21 prior to or adjunct to CGM in persons with longstanding T1DM

and impaired awareness of hypoglycaemia, could help differentiate between those in

need of continuous versus short-term or intermittent use of CGM.22

Recent developments in diabetes technology may further improve patient

convenience and efficacy of automated glucose measurement techniques and feedback.

For example, flash glucose monitoring23 and the first FDA approved hybrid closed-loop

system24 have showed promising results regarding glycaemic outcomes. Future studies

investigating peoples lived experiences with these new technologies are warranted, to

ensure successful integration into the daily lives of persons with diabetes and provide

evidence-based guidance for clinicians. As long as human behaviour is involved in

managing technology, individual attitudes, coping and self-management styles are

important and should be considered in use and education of new technology.

Our qualitative study has some limitations that deserve to be mentioned. First,

selection bias might have occurred towards those with a more positive attitude

towards diabetes technology and CGM in particular. We should therefore be cautious in

generalising our findings. Furthermore, during the trial, support from the health care

133

team might have been more intense compared with care in routine clinical practice.

Another limitation is that we asked participants retrospectively to look back on their

expectations and experiences rather than during the study. Due to the cross-over

design time since using and returning the sensor were different, although we have no

indication that the sequence (SMBG/CGM vs. CGM/SMBG) had an impact on the results.

We did not check our results with the participants (member checking), which could

have been an additional validation of our findings. However, we are confident that the

results are valid and indeed largely concord with previous studies.

In conclusion, our qualitative study in adults with T1DM and impaired awareness of

hypoglycaemia shows that CGM adds to their sense of control and well-being confirming

previous research. For this specific population at risk for recurrent hypoglycaemia the

primary benefits of CGM are feelings of safety and relief rather than convenience.

Active and passive behavioural responses to CGM were reported, indicating differential

use of CGM based on individual attitudes and preferred coping styles. Future studies

should explore differential use of CGM in this population in the context of active and

passive self-management styles.

Ch

apte

r 6

134






971-8.





937-48.















36(12): 4160-2.









1240-7.


SJ, et al. Continuous glucose

monitoring for patients with type 1

diabetes and impaired awareness

of hypoglycaemia (IN CONTROL): a

randomised, open-label, crossover

trial. Lancet Diabetes Endocrinol

2016.







28(10): 1158-67.

8. Kubiak T, Mann CG, Barnard KC,

Heinemann L. Psychosocial Aspects

of Continuous Glucose Monitoring:

Connecting to the Patients'

Experience. J Diabetes Sci Technol

2016; 10(4): 859-63.

9. Polonsky WH, Hessler D. What are

the quality of life-related benefits

and losses associated with real-time

continuous glucose monitoring? A

survey of current users. Diabetes

Technol Ther 2013; 15(4): 295-301.





2010; 27(9): 1060-5.




REFERENCES

135



2015; 38(4): 544-50.

12. Ritholz MD, Beste M, Edwards

SS, Beverly EA, Atakov-Castillo A,

Wolpert HA. Impact of continuous

glucose monitoring on diabetes

management and marital

relationships of adults with Type

1 diabetes and their spouses: a

qualitative study. Diabet Med 2014;

31(1): 47-54.

13. Juvenile Diabetes Research


Monitoring Study G, Beck RW,

Lawrence JM, et al. Quality-of-life

measures in children and adults with

type 1 diabetes: Juvenile Diabetes

Research Foundation Continuous

Glucose Monitoring randomized

trial. Diabetes Care 2010; 33(10):

2175-7.

14. Braun V, Clarke V. Using thematic

analysis in psychology. Qualitative

Research in Psychology 2006;

3:77-101.

15. Russell CK, Gregory DM. Evaluation

of qualitative research studies. Evid

Based Nurs 2003; 6(2): 36-40.

16. Tong A, Sainsbury P, Craig J.

Consolidated criteria for reporting

qualitative research (COREQ): a

32-item checklist for interviews and

focus groups. Int J Qual Health Care

2007; 19(6): 349-57.

17. Borges U, Jr., Kubiak T. Continuous

Glucose Monitoring in Type 1

Diabetes: Human Factors and Usage.

J Diabetes Sci Technol 2016.

18. Schmidt S, Duun-Henriksen AK,

Norgaard K. Psychosocial factors

and adherence to continuous glucose

monitoring in type 1 diabetes. J

Diabetes Sci Technol 2012; 6(4):

986-7.

19. Litchman ML, Allen NA. Real-Time


Facilitates Feelings of Safety in

Older Adults With Type 1 Diabetes:

A Qualitative Study. J Diabetes Sci

Technol 2017: 1932296817702657.





Diabetes Care 2001; 24(4): 637-42.

21. Rondags SM, de Wit M, Twisk

JW, Snoek FJ. Effectiveness

of HypoAware, a Brief Partly

Web-Based Psychoeducational

Intervention for Adults With Type 1

and Insulin-Treated Type 2 Diabetes



Diabetes Care 2016; 39:2190-2196.







38(8): 1592-609.

Ch

apte

r 6

136

23. Bolinder J, Antuna R, Geelhoed-

Duijvestijn P, Kroger J, Weitgasser

R. Novel glucose-sensing technology

and hypoglycaemia in type 1

diabetes: a multicentre, non-masked,

randomised controlled trial. Lancet

2016; 388(10057): 2254-63.

24. Bergenstal RM, Garg S, Weinzimer

SA, et al. Safety of a Hybrid

Closed-Loop Insulin Delivery System

in Patients With Type 1 Diabetes.

JAMA 2016; 316(13): 1407-8.

Journal of Diabetes and its Complications 2017

Cornelis A.J. van Beers, Martine G. Caris, J. Hans DeVries, Erik H. Serné

7

The relation between

HbA1c and hypoglycaemia

revisited; a secondary analysis

from an intervention trial in

patients with type 1 diabetes

and impaired awareness of

hypoglycaemia

Ch

apte

r 7

140

ABSTRACT

Aims

We aimed to re-assess the previously shown but recently disputed association

between HbA1c and severe hypoglycemia.

Methods

52 Patients with T1D and IAH participated in an earlier reported randomized,

crossover trial with two 16-week intervention periods comparing continuous glucose

monitoring (CGM) with self-monitoring of blood glucose (SMBG). In this previous

study, time spent in normoglycemia (the primary outcome), was improved by 9.6%

(p<0.0001). We performed post-hoc analyses using a zero-inflated Poisson regression

model to assess the relationship between severe hypoglycemia and HbA1c, glucose

variability and duration of diabetes.

Results

During SMBG use, HbA1c and the number of severe hypoglycemic events were

negatively associated (OR 0.20 [95% CI 0.09 to 0.44]). During CGM use, this relationship

showed an odds ratio of 0.65 (95% CI 0.42 to 1.01). There was no significant relationship

between glucose variability or duration of diabetes and severe hypoglycemia.

Conclusions

In patients with T1D and IAH, treated with standard SMBG, a negative association

exists between HbA1c and the number of severe hypoglycemic events. Thus, reaching

target HbA1c values still comes with a higher risk of severe hypoglycemia. CGM

weakens this association, suggesting CGM enables patients to reach their target HbA1c

more safely.

Keywords

Type 1 diabetes, hypoglycemia, severe hypoglycemia, impaired awareness of

hypoglycemia, HbA1c

141

INTRODUCTION

In patients with type 1 diabetes (T1D), hypoglycemia is the main limiting factor in

achieving satisfactory glucose control.1 In the DCCT, a quadratic relationship between

HbA1c and the frequency of severe hypoglycemia was seen, with the risk of severe

hypoglycemia clearly increasing as HbA1c decreased.2,3 This was confirmed by the

EURODIAB IDDM Complications Study, in which 40.1% of patients with an HbA1c

<5.4% were affected by severe hypoglycemia compared with 24.3% of patients with

an HbA1c of >7.8%.4 However, more recent analyses from the T1D Exchange clinic

registry do not show a relationship between HbA1c and severe hypoglycemia,5 neither

in adults6,7 nor in children.8 These reports suggested that advances in the management

of T1D over the last 20 years with structured education, rapid- and long-acting insulin

analogues, insulin pumps, and continuous glucose monitoring (CGM) have weakened

the association between HbA1c and severe hypoglycemia. Importantly, several other

studies indicate that other factors, such as hypoglycemia awareness status9,10 and

duration of diabetes10,11 might be more important in identifying individuals prone to

severe hypoglycemia.

We aimed to re-assess the previously shown but recently disputed association

between HbA1c and severe hypoglycemia, using data from a recently reported

randomized, crossover trial (IN CONTROL) comparing the effect of CGM with self-mo-

nitoring of blood glucose (SMBG) in adult patients with T1D and impaired awareness

of hypoglycemia (IAH).12 We also assessed the relationship between CGM-assessed

hypoglycemia and HbA1c. Subsequently, we investigated whether CGM modifies this

association. Furthermore, we aimed to assess whether duration of diabetes is associated

with severe hypoglycemia within this population of patients with long-standing T1D.

In addition, since re-analyses of the DCCT showed that glycemic variability also relates

to hypoglycemic risk,3 we evaluated the association between short-term glucose

variability and severe hypoglycemia.

Ch

apte

r 7

142

SUBJECTS, MATERIALS AND METHODS

Study design and subjects

The IN CONTROL trial was a two-center, randomized, open-label, crossover trial

with a 12-week wash-out period in between 16-week intervention periods, comparing

CGM with SMBG. A detailed description of the study protocol has been published

previously.13 In brief, 52 adult (mean age 48.6 years ± 11.6) patients with T1D and IAH

as confirmed by a Gold score9 ≥4 and treated with either continuous subcutaneous

insulin infusion (CSII) or multiple daily injections (MDI) were randomized to the CGM/

SMBG (n=26) or the SMBG/CGM (n=26) sequence. During the CGM phase, patients

used a real-time CGM device consisting of the Paradigm Veo system with a MiniLink

transmitter and the Enlite glucose sensor (Medtronic, Northridge, CA, USA). During the

SMBG phase, patients used a masked CGM system consisting of an iPro2 CGM and an

Enlite glucose sensor (Medtronic, Northridge, CA, USA). This masked CGM system did

not display real-time CGM data or glucose trends nor allowed alarms to be set. During

both intervention periods, patients were instructed to use the CGM system continuously

throughout the 16-week intervention period. During both periods, patients attended

monthly follow-up visits to review glucose data, based on CGM data during the CGM

phase and SMBG data during the SMBG phase, and adjust therapy accordingly. These

visits were followed by telephone consultations two weeks after each follow-up visit,

comprising inquiry after adverse events, episodes of (severe) hypoglycemia, study

device use and related technical issues, and medication changes. The protocol was

approved by the medical ethics committee of the VU University Medical Center.

Written informed consent was obtained from all patients. This trial is registered with

ClinicalTrial.gov, number NCT01787903.

Measurements

Primary endpoint of the trial was the difference time spent in normoglycemia (4

– 10 mmol/L) between the total 16-week CGM and SMBG phase. Secondary outcomes

included time spent ≤3.9 mmol/L and time spent >10 mmol/L, HbA1c, the number of

severe hypoglycemic events (defined as a hypoglycemic event requiring third party

assistance), the proportion of patients affected by at least one severe hypoglycemic

event, CGM-assessed hypoglycemia (e.g. the frequency of CGM-assessed hypoglycemic

events <2.8 mmol/L per week and AUC ≤3.9 mmol/L), and short term glucose variability

(e.g. within-day coefficient of variation [%CV]). Since within-day %CV has a relatively

strong relation with hypoglycemia rate, within-day %CV was used as the only metric of

short term glucose variability, also to limit the risk of overtesting.14,15 The within-day

%CV was defined as the average of all daily %CV’s. The daily %CV was calculated as

‘(100 × standard deviation of all daily glucose values) divided by the mean of all daily

143

glucose values’.16 The CGM-assessed outcomes were calculated over the total 16-week

intervention phase. The primary and secondary outcomes of the trial have been

reported previously.12 In brief, time spent in normoglycemia during the CGM phase

was higher than time spent in normoglycemia during the SMBG phase (65.0% vs 55.4%;

Δ 9.6 [95% CI 8.0 to 11.2]). In addition, the number of severe hypoglycemic events was

lower during the CGM phase (14 events versus 34 events, p=0.033). HbA1c at the end of

each period was similar (CGM 7.3% [56 mmol/mol] and SMBG 7.3% [56 mmol/mol], Δ

0.0% [95% CI -0.1 to 0.2]).

Statistical analysis

All analyses are post-hoc analyses of data from our previously published IN CONTROL

trial. We used a zero-inflated Poisson regression model, stratified for intervention

phase, i.e. SMBG and CGM, to assess the relationship between number of severe

hypoglycemic events during the 16-week intervention phase and HbA1c, within-day

%CV and duration of diabetes, respectively. We used linear regression to assess the

relationship between the CGM-assessed hypoglycemia and HbA1c, within-day %CV

and duration of diabetes, respectively. A quadratic regression provided the best fit

for the distribution of the data. All models were stratified for intervention phase, i.e.

SMBG and CGM. The frequency of CGM-assessed hypoglycemic events <2.8 mmol/L per

week was compared between patients affected by at least one severe hypoglycemic

event and non-affected patients using an independent samples t-test. STATA/SE 14.1 for

Windows was used for the zero-inflated Poisson regression analyses, all other analyses

were conducted using SPSS 22.0 for Windows (IBM SPSS Inc., Chicago, IL, USA).

Ch

apte

r 7

144

RESULTS

Baseline characteristics are shown in Table 1. The intention-to-treat population

consisted of all 52 randomized patients. The mean age of the population was 48.6 ±

11.6 years, 24 (46%) were female, median duration of diabetes was 30.5 years (IQR

18.5 - 40.8), with a minimum duration of 4.1 years and a maximum duration of 55.5

years. 23 patients (44 %) were treated with CSII, and mean HbA1c was 7.5% ± 0.8% (58

± 8.7 mmol/mol). Of the 29 patients treated with MDI, 28 (96.6%) were treated with a

combination of a long-acting insulin analogue and rapid-acting insulin. One patient

was treated with a combination of NPH-insulin and rapid-acting insulin.

Intention-to-treat population (n=52)

Female 24 (46.2%)

Age (years) 48.6 (11.6)

Diabetes duration (years) 30.5 (18.5-40.8)

HbA1c (%) 7.5 (0.8)

HbA1c (mmol/mol) 58.1 (8.7)

Gold score9 5.4 (0.7)

Insulin treatment modality, MDI, 29 (55.8%)

Long-acting analogues

Detemir 8 (27.6%)

Glargine 20 (69.0%)

Intermediate-acting analogues

NPH 1 (3.4%)

Rapid-acting insulins

Aspart 26 (89.7%)

Lispro 3 (10.3%)

Insulin treatment modality, CSII 23 (44.2%)

Aspart 6 (26.1%)

Lispro 15 (65.2%)

Regular human insulin 2 (8.7%)

Duration of CSII therapy (years) 10.2 (4.7, 18.7)

Self-reported daily home glucose-meter readings (no./day) 5.0 (4.0, 6.0)

Data are n (%), mean (SD), or median (25%, 75%). MDI: multiple daily injections; CSII: continuous subcutaneous insulin

infusion.

Table 1. Baseline characteristics

145

During the SMBG phase, 34 severe hypoglycemic events occurred in 18 (35%) patients.

During the CGM phase, 14 severe hypoglycemic events occurred in 10 (19%) patients.

Zero-inflated Poisson regression analysis showed a significant negative association

between HbA1c and severe hypoglycemic events during the SMBG phase (OR 0.20 [95%

CI 0.09 to 0.44]; Figure 1). During the CGM phase, this relationship showed an OR of

0.65 (95% CI 0.42 to 1.01). No significant relationship was seen between within-day

%CV and severe hypoglycemia during the SMBG phase (OR 1.00 [95% CI 0.78 to 1.3])

or during the CGM phase (OR 1.07 [95% CI 0.95 to 1.22]. Duration of diabetes was also

not associated with the number of severe hypoglycemic events, during standard SMBG

treatment (OR 1.02 [95% CI 0.97 to 1.07]) or CGM (OR 1.00 [95% CI 0.95 to 1.04].

After log transformation of the frequency of CGM-assessed hypoglycemic events

<2.8 mmol/L, linear regression analysis showed a significant association between

HbA1c and the frequency of CGM-assessed hypoglycemic events <2.8 mmol/L during

both intervention phases (SMBG: OR 0.61 [95% CI 0.50 to 0.76]; CGM: OR 0.56 [95% CI

0.34 to 0.91]. In addition, HbA1c was significantly associated with the AUC ≤3.9 mmol/L

during both intervention phases (SMBG OR 0.55 [95% CI 0.46 to 0.67]; CGM: OR 0.54

6

4

2

0

Mea

n n

um

ber

of

seve

re h

ypog

lyce

mic

eve

nts

SMBG

CGM

This figure shows the relationship between HbA1c values (%) and mean number of severe hypoglycaemic events, based on the zero-inflated Poisson regression model, during both the SMBG phase (solid black line) and the CGM phase (solid grey line). Dashed lines represent 95% confidence intervals.

Figure 1. Relationship between HbA1c and the mean number of severe hypoglycaemic events

6,0 6,5 7,0 7,5 8,0 8,5 9,0

HB1C (%)

Ch

apte

r 7

146

[95% CI 0.37 to 0.76]. Within-day %CV was significantly associated with the frequency

of CGM-assessed hypoglycemic events <2.8 mmol/L during both phases (SMBG: OR 1.12

[95% CI 1.08 to 1.15]; CGM: OR 1.24 [95%CI 1.15 to 1.33]) and with the AUC ≤3.9 mmol/L

during both phases (SMBG: OR 1.12 [95% CI 1.08 to 1.16]; CGM: OR 1.20 [95% CI 1.30

to 1.27]). Duration of diabetes was not significantly associated with the frequency of

CGM-assessed hypoglycemic events <2.8 mmol/L during either phase (SMBG: OR 0.99

[95% CI 0.98 to 1.01]; CGM: OR 0.99 [95% CI 0.96 to 1.01]) or with the AUC ≤3.9 mmol/L

during either phase (SMBG: OR 0.99 [95% CI 0.98 to 1.01]; CGM: OR 0.98[95% CI 0.96 to

1.00]).

In patients affected by at least one severe hypoglycemic event, the median frequency

of CGM-assessed hypoglycemic events <2.8 mmol/L was 4.5 events per week (IQR 2.9-

5.5) and 2.9 events per week (IQR 1.8–4.8) in patients not affected (p=0.027).

147

DISCUSSION

In this high-risk population of patients with long-standing type 1 diabetes and

impaired awareness of hypoglycemia, HbA1c is negatively associated with the number

of severe hypoglycemic events as well as CGM-assessed hypoglycemia, despite use of

rapid- and long-acting insulin analogues, insulin pumps, and a median of five SMBG

measurements per day. CGM significantly weakens the association between HbA1c

and severe hypoglycemia, but not between HbA1c and hypoglycemic events <2.8

mmol/L. Glycemic variability, defined as with-in day %CV, was not associated with the

number of severe hypoglycemic events. Also, within this population of patients with

long-standing diabetes, duration of diabetes was not associated with the number of

severe hypoglycemic events.

After the DCCT, various studies demonstrated variable relationships between HbA1c

and severe hypoglycemia.3-8,17 However, this relationship has not been assessed

specifically in adult patients with T1D and IAH and it is unclear whether this relationship

can be modified by continuous glucose monitoring (CGM). In adult patients with T1D

and IAH, we show a negative relationship between HbA1c and severe hypoglycemia as

well as documented hypoglycemic events <2.8 mmol/L. Interestingly, CGM significantly

weakens the association between HbA1c and severe hypoglycemia, but not between

HbA1c and hypoglycemic events <2.8 mmol/L. Thus, it seems that CGM protects against

severe hypoglycemic events without altering the relationship between HbA1c and

hypoglycemic events <2.8 mmol/L. CGM therefore might function as an emergency

brake enabling patients to reach their target HbA1c values more safely.

Our results are discordant with the results of the T1D Exchange, which did not show

a straightforward relationship between severe hypoglycemia and HbA1c.6-8 The T1D

Exchange reports cross-sectional observations analyzing a general T1D population. The

T1D Exchange 6,7 and other trials have shown that in these general T1D populations,

factors other than HbA1c such as hypoglycemia awareness status9,10 and duration

of diabetes10,11 might be more important in identifying individuals prone to severe

hypoglycemia. However, all participants of the IN CONTROL trial were impaired aware

of hypoglycemia and had diabetes for a long time. Within this selected population, and

the setting of a randomized trial, HbA1c was strongly associated with hypoglycemia.

It is well known that HbA1c does not provide a complete picture of an individuals’

risk of hypoglycemia. Re-analyses of the DCCT showed that roughly 5% of the difference

in frequency of hypoglycemia could be attributed to HbA1c.3 Glucose variability or

Ch

apte

r 7

148

dispersion, especially when expressed as coefficient of variation, has also been

shown predictive of severe hypoglycemia.14,18,19 Our data however do not confirm this

association. Surprisingly, we did show a strong association between glucose variability

and CGM-assessed hypoglycemia. This discrepancy might suggest that CGM-assessed

hypoglycemia, also with a low cutoff value, is not suitable as a proxy for severe

hypoglycemia in clinical research. Future research should further explore this.

Previously, a clear difference has been shown in the proportion of patients affected

by severe hypoglycemia between patients with short- (<5 years) and long-standing (>15

years) diabetes.11 This has been attributed in part to more heavily affected counter-re-

gulatory mechanisms in the latter.20 In our trial, duration of diabetes was not associated

with neither severe hypoglycemia, nor the frequency of CGM-assessed hypoglycemic

events <2.8 mmol/L (calculated over 16 weeks of continuous glucose data), which is to

be expected from our sample with a median duration of diabetes of 30.5 years (IQR

18.5- 40.8), and a minimum duration of 4.1 years. In addition, impaired awareness of

hypoglycemia and defective counter-regulatory mechanism probably share a similar

pathophysiology.21 It is therefore likely that counter-regulatory mechanism were

affected in all patients in our trial, explaining the absent association between the

duration of diabetes and hypoglycemia in this patient population.

With regard to the external validity, the study population of the IN CONTROL trial

seems to reflect a typical, real-life population of adult patients with T1D and IAH, based

on the mean age (48.6 years [SD 11.6]), the median diabetes duration (30.5 years [IQR

18.5 to 40.8]), the mean baseline HbA1c value (7.5% [SD 0.8]), the proportion of patients

treated with CSII (23 patients [44.2%]), and the insulins used by the patients.22-24

We acknowledge that this trial was not primarily designed to investigate associations

between severe hypoglycemia and HbA1c, glucose variability and duration of diabetes.

Our results are limited by the fact that study duration was relatively short (two times

16 weeks), the number of patients and the number of severe hypoglycemic events

were relatively small, the latter especially during CGM use.

In conclusion, in typical adult patients with type 1 diabetes and impaired awareness

of hypoglycemia, treated with standard SMBG, there is an inverse association between

HbA1c and both the number of severe hypoglycemic events and CGM-assessed

hypoglycemia. Thus, reaching target HbA1c values comes with a higher risk of severe

hypoglycemia and setting a higher HbA1c target for people with impaired awareness of

hypoglycemia remains a valid strategy. CGM weakens the association between HbA1c

and severe hypoglycemia and therefore treating patients with CGM enables them

149

to reach their target HbA1c values more safely. For patients with T1D and impaired

awareness of hypoglycemia, the major barrier for achieving normoglycemia still

stands.

Ch

apte

r 7

150



management of Type I and

Type II diabetes. Diabetologia.

2002;45:937-48.







diabetes mellitus. The Diabetes

Control and Complications Trial

Research Group. N Engl J Med.

1993;329:977-86.



Hypoglycemia in the Diabetes

Control and Complications Trial.

Diabetes. 1997;46:271-86.

4. EUROBIAB IDDM Complications

Study Group. Microvascular and

acute complications in IDDM

patients: the EURODIAB IDDM

Complications Study. Diabetologia.

1994;37:278-85.

5. Miller KM, Foster NC, Beck RW,

Bergenstal RM, DuBose SN, DiMeglio

LA, et al. Current state of type 1

diabetes treatment in the U.S.:

updated data from the T1D Exchange

clinic registry. Diabetes Care.

2015;38:971-8.

6. Weinstock RS, Xing D, Maahs DM,

Michels A, Rickels MR, Peters AL,

et al. Severe hypoglycemia and

diabetic ketoacidosis in adults with

type 1 diabetes: results from the

T1D Exchange clinic registry. J Clin

Endocrinol Metab. 2013;98:3411-9.

7. Simmons JH, Chen V, Miller KM,

McGill JB, Bergenstal RM, Goland RS,

et al. Differences in the management

of type 1 diabetes among adults

under excellent control compared

with those under poor control in

the T1D Exchange Clinic Registry.

Diabetes Care. 2013;36:3573-7.

8. Campbell MS, Schatz DA, Chen V,

Wong JC, Steck A, Tamborlane WV,

et al. A contrast between children

and adolescents with excellent and

poor control: the T1D Exchange

clinic registry experience. Pediatr

Diabetes. 2014;15:110-7.





hypoglycemia. Diabetes Care.

1994;17:697-703.

10. Pedersen-Bjergaard U, Pramming S,

Heller SR, Wallace TM, Rasmussen

AK, Jorgensen HV, et al. Severe




Metab Res Rev. 2004;20:479-86.

REFERENCES

151





Diabetologia. 2007;50:1140-7.


SJ, Smits MM, Geelhoed-Duijvestijn

PH, Kramer MH, et al. Continuous

glucose monitoring for patients

with type 1 diabetes and impaired


(IN CONTROL): a randomised,

open-label, crossover trial. Lancet

Diabetes Endocrinol. 2016;4:893-902.

13. van Beers CA, Kleijer SJ, Serne EH,

Geelhoed-Duijvestijn PH, Snoek

FJ, Kramer MH, et al. Design and

rationale of the IN CONTROL trial:

the effects of real-time continuous

glucose monitoring on glycemia and




Endocr Disord. 2015;15:42.

14. Qu Y, Jacober SJ, Zhang Q, Wolka LL,

DeVries JH. Rate of hypoglycemia in

insulin-treated patients with type

2 diabetes can be predicted from

glycemic variability data. Diabetes

Technol Ther. 2012;14:1008-12.



measure it. Diabetes. 2013;62:1405-8.

16. Rodbard D. Interpretation of

continuous glucose monitoring data:

glycemic variability and quality of

glycemic control. Diabetes Technol

Ther. 2009;11 Suppl 1:S55-S67.

17. Karges B, Rosenbauer J, Kapellen T,

Wagner VM, Schober E, Karges W, et

al. Hemoglobin A1c Levels and risk

of severe hypoglycemia in children

and young adults with type 1

diabetes from Germany and Austria:

a trend analysis in a cohort of 37,539

patients between 1995 and 2012.

PLoS Med. 2014;11:e1001742.

18. Cox DJ, Kovatchev BP, Julian DM,

Gonder-Frederick LA, Polonsky WH,

Schlundt DG, et al. Frequency of

severe hypoglycemia in insulin-de-

pendent diabetes mellitus can be

predicted from self-monitoring

blood glucose data. J Clin Endocrinol

Metab. 1994;79:1659-62.

19. Kilpatrick ES, Rigby AS, Goode K,

Atkin SL. Relating mean blood

glucose and glucose variability to

the risk of multiple episodes of

hypoglycaemia in type 1 diabetes.

Diabetologia. 2007;50:2553-61.

20. Cryer PE. Mechanisms of sympatho-

adrenal failure and hypoglycemia

in diabetes. J Clin Invest.

2006;116:1470-3.

21. Cryer PE. Diverse causes of hypogly-

cemia-associated autonomic

failure in diabetes. N Engl J Med.

2004;350:2272-9.






Ch

apte

r 7

152




Diabet Med. 2010;27:666-72.


L, Gallen G, Pender S, Brackenridge

A, et al. Real-time continuous



hypoglycemia-unaware patients

with type 1 diabetes. Diabetes Care.

2013;36:4160-2.








clinical trial. JAMA. 2013;310:1240-7.

8

Summary & General

Discussion

Ch

apte

r 8

156

In this thesis, the effect of continuous glucose monitoring on glycaemic control

and psychological outcomes in patients with type 1 diabetes and impaired awareness

of hypoglycaemia is addressed. In addition, the potential modifying effects of

psychological distress, the patients’ experiences of continuous glucose monitoring, and

the relationship between HbA1c and hypoglycaemia are explored.

In this final chapter, the main findings will be summarised and discussed with

special emphasis on the possible implications for clinical practice.

Summary and general discussion

Gap in the evidence

In the mid-20th century, the clinical syndrome of impaired awareness of hypoglycaemia

(IAH) in diabetes was well described by Lawrence.1 Specifically, the absence of

forewarning adrenergic symptoms made it difficult to avoid the consequences of

neuroglycopaenia, such as confusion, seizures, coma, and even death.2 Although there

is still debate over whether this is to be considered a maladaptive or adaptive response

of the brain to repeated events,3 the risks from severe hypoglycaemia are a serious

burden for roughly 25% of adults with type 1 diabetes (T1D) who suffer from IAH.

Historically, patient education and adherence to diabetes management strategies have

been the mainstays of hypoglycaemia prevention, but technological developments have

provided additional approaches.4 Continuous subcutaneous insulin infusion (CSII), for

example, might be associated with less risk of severe hypoglycaemia compared with

multiple daily insulin injections (MDI),5 but assessing this risk among patients with IAH

is difficult without studies designed specifically to look at this group.

Before the IN CONTROL study, described in the present thesis, only a few trials

in patients with T1D and IAH have examined the effects of continuous glucose

monitoring (CGM). As described in Chapter 2, the Cochrane collaboration found

no difference in the incidence rates of severe hypoglycaemia between CGM and

self-monitoring of blood glucose (SMBG), when summarising data up to June, 2011,

and analysing studies where (unfortunately) patients with IAH had mostly been

excluded.6 Nevertheless, in an observational study one year later, Choudhary et al.

demonstrated a reduction of severe hypoglycaemia in patients with IAH7 supported

by CGM, reinforcing the need for targeted randomised studies in patients with IAH. A

randomised controlled trial using CGM with low-glucose suspend seemed to confirm a

benefit by demonstrating a reduction of severe hypoglycaemia in patients with IAH.8

However, the population studied was relatively young (mean age 18.6 years) and the

reduction of severe hypoglycaemia lost statistical significance when two outliers in the

157

youngest age groups were excluded from the analyses. Since patients with T1D and

IAH are usually older than 40 years and have more than 25 years of diabetes duration,9

this study left important questions unanswered. It is possible that patients with longer

duration of diabetes, who may have associated problems such as lipohypertrophy,

anti-insulin antibodies and more entrenched health beliefs that may predispose to

severe hypoglycaemia, may not be able to demonstrate similar reductions in severe

hypoglycaemia events. This was addressed in another recent randomised controlled

trial specifically focusing on adults with T1D and IAH.10 This trial reported improved

hypoglycaemia awareness and glycaemic control from baseline to endpoint (24 weeks)

offering extensive patient guidance that included weekly contact, monthly follow-up

visits, and use of a bolus calculator to determine the insulin dose, irrespective of insulin

pump use. However, no added benefit of CGM was shown. Importantly, sensors were

used a median of 57% of the time in this study; only 17 of the 42 individuals achieved

80% sensor usage threshold, which is often considered the frequency required for

meaningful benefit. Therefore, whether CGM adds any benefit in patients with T1D

and IAH was still unknown.

Effects of CGM on glycaemic control and severe hypoglycaemia

Chapter 4 of this thesis demonstrated that, in adults with T1D and IAH, a 16-week

intervention with CGM (without low-glucose suspension) significantly improved time

spent in normoglycaemia, with less time spent in hypoglycaemia and hyperglycaemia,

compared with SMBG. In addition, CGM decreased the frequency of CGM-derived

hypoglycaemia and severe hypoglycaemia (i.e. requiring third party assistance).

Interestingly, the lower the biochemical cut-off for hypoglycaemia was set, the larger

the relative reduction in hypoglycaemia was observed with CGM, which might suggest

that patients in the IN CONTROL trial took action only when their glucose level was

already ≤3·9 mmol/L, or perhaps defined a lower threshold as relevant for self-treating

hypoglycaemia. Importantly, CGM reduced the frequency of hypoglycaemic episodes

<2·8 mmol/L by 44.0%, which are prone to cause cognitive dysfunction as a result of

neuroglycopaenia.11 Also, the proportion of patients affected by severe hypoglycaemia

during the 16-week intervention periods was lower during CGM than during SMBG.

This effect occurred without increasing HbA1c, which is often a price to pay when

trying to avoid hypoglycaemia. Also, in Chapter 7, we showed that in this high-risk

population of patients with long-standing T1D and IAH, HbA1c is negatively associated

with the number of severe hypoglycemic events while using standard SMBG, despite

use of rapid- and long-acting insulin analogues, insulin pumps, and a median of

five SMBG measurements per day. CGM significantly weakens this association and

therefore treating patients with CGM enables them to reach their target HbA1c values

more safely. These results support our hypothesis that CGM is a valuable tool in the

Ch

apte

r 8

158

(safe) treatment of adult patients with T1D and IAH.

Our findings indicate that CGM must be used continuously, not intermittently,

by patients in order to be effective. As shown in Chapter 4, the improvements

in CGM-derived glycaemic outcomes (i.e. time spent in normoglycaemia, AUC in

hypoglycaemia) and reduction in severe hypoglycaemia were found in patients with

a median sensor usage of 89.4% (IQR 80.8 to 95.5). Also, the difference in time spent

in normoglycaemia between SMBG and CGM shown in Chapter 4 is comparable to

the difference between patients using CGM <50% and ≥50% of the time as reported in

another intervention trial performed in patients with IAH.10 This is also in line with

previous trials that consequently showed that CGM is only effective when used at least

60% of the time.12-14

Effects of CGM on glucose variability

Patients with T1D face an optimisation problem: reducing HbA1c values while

simultaneously avoiding hypoglycaemia. It seems common sense that bringing

down average glycaemic values is only possible if glucose variability is constrained,

otherwise the blood glucose fluctuations would inevitably enter the hypoglycaemic

range. Surprisingly, although we demonstrated a strong association between glucose

variability and the frequency of CGM-assessed hypoglycaemic events <2.8 mmol/L

in Chapter 7, we could not confirm the association between severe hypoglycaemia

and glucose variability in adults with T1D and IAH. However, re-analyses of the

DCCT data has established that not only HbA1c, but also mean blood glucose level,

and glucose variability measurements each have an independent role in determining

an individual's risk of hypoglycaemia in T1D,15 suggesting that all three aspects of

glycaemic assessment should thus be considered in patients in whom hypoglycaemia

is a real or potential problem. Also, several other trials showed that glucose variability

or dispersion, especially when expressed as coefficient of variation, is predictive of

severe hypoglycemia.16,17 Future trials should further explore to what extent glucose

variability is a determinant of severe, clinically meaningful hypoglycaemia. In

Chapter 4, we showed that CGM improved short term (within-day and between-days)

glucose variability, without change in HbA1c. Interestingly, the improved CGM-derived

measures of glycaemic control (i.e. time spent in target and glucose variability) were

indeed paralleled by clinically relevant reduction in severe hypoglycaemia.

Long-term learning effects of CGM

A long-term learning effect of CGM would be extremely beneficial since it would

enable clinicians to treat and guide patients with CGM for a short term, and stop CGM as

soon as the patient had learned to avoid hypoglycaemia. This would reduce high costs

159

that go hand in hand with CGM and therefore might result in CGM to be used by more

patients (for a short term). Also, a learning effect of CGM could be expected, considering

the ability of patients to e.g. recognize patterns in otherwise undetected hypo- or

hyperglycaemic events and adjust self-management accordingly. Unfortunately, in

Chapter 4, the data seem to confirm that CGM does not have a long-term learning

effect,18 because withdrawal of CGM resulted in time spent in normoglycaemia to

fall back to baseline values after 12 weeks. We acknowledge that the IN CONTROL

trial was not primarily designed to demonstrate a learning effect, that CGM was not

accompanied by a structured educational program, and that we did not formally ask

our patients to learn from the real-time and retrospectively assessed CGM-data. But,

during the trial patients did attend monthly follow-up visits during which ongoing

education (e.g. regarding timing and frequency of mealtime boluses, insulin stacking,

alcohol use, exercise, incorporating real-time CGM-data into patients’ self-management

etc.) was provided by the research physicians. The research physicians and patients

discussed the uploaded CGM-data and previous self-management decisions in detail

and made therapy changes together. Even with this frequent guidance and shared

decision making, the beneficial effect of CGM completely disappeared after 12 weeks.

We should therefore be cautious in expecting relevant long-term effects of CGM.

Formal evidence investigating long-term learning effects of CGM, when combined with

a structured education program such as HypoAware,19 is however lacking and trials

providing this evidence should be endorsed.

Effects of CGM on hypoglycaemia awareness

Our scepticism regarding long-term effects of CGM is supported by our finding

that CGM, used as stand-alone intervention, does not seem to improve hypoglycaemia

awareness in a clinically relevant manner after 16 weeks. In Chapter 4 we noted

no relevant differences in self-reported hypoglycaemia awareness scores, with no

relevant between-group differences in 16-week hypoglycaemia awareness scores or

significant change in hypoglycaemia awareness scores from baseline to endpoint. In

2011, a clamp study by Ly et al. showed that 4 weeks of CGM improved epinephrine

responses in young T1D patients with IAH, suggesting that IAH can be restored

in adolescents by using CGM,20 but this finding was not supported by a larger trial

performed by the same study group.8 An observational trial by Choudhary et al.

did not demonstrate a difference in self-reported hypoglycaemia awareness scores

before and after CGM.7 Furthermore, although some randomised controlled trials

assessing the effect of CGM on hypoglycaemia awareness did demonstrate a significant

improvement in self-reported hypoglycaemia awareness during these trials, these

trials failed to demonstrate any between-group differences between SMBG and

CGM.8,10 Small studies have shown that meticulous avoidance of hypoglycaemia is

Ch

apte

r 8

160

required to correct hypoglycaemia-induced counter-regulatory defects which may

improve hypoglycaemia awareness.21 As shown in Chapter 4, CGM does not prevent all

hypoglycaemia, but only decreases the duration and depth of hypoglycaemic episodes.

More rigorous avoidance of hypoglycaemic events for a longer period of time might be

needed to improve hypoglycaemia awareness.

Effects of CGM on quality of life and patients’ perspectives

As shown in Chapter 4 and Chapter 5, CGM enables patients to worry less about

hypoglycaemia, as CGM improves glycaemic control and significantly lowers the actual

risk of severe hypoglycaemia. CGM increases treatment satisfaction.12 but surprisingly,

does not seem to have a profound measurable effect on other (diabetes-specific)

markers of quality of life,22,23 such as diabetes-related distress, emotional well-being

and diabetes self-efficacy (Chapter 4). This could be explained by a high level of quality

of life at baseline in these studies suggesting a ceiling effect. A recent CGM trial in

adults with T1D who use multiple daily insulin injections however did demonstrate

that CGM might contribute to improvement in diabetes-specific markers of quality of

life, but not to quality of life measures not specific to diabetes.24 Data therefore remain

inconclusive about the influence of CGM on quality of life measures in patients with

T1D.

However, studies using a qualitative research approach consistently demonstrate

an overall positive impact of CGM on quality of life.25 In Chapter 6, to further our

understanding of the individual use and experience of CGM in patients with T1D and

IAH, a qualitative study supplementary to the IN CONTROL trial was conducted, using

semi-structured interviews. We demonstrated that the positive experiences with CGM

clearly outweigh the negative. The benefits experienced with CGM in patients T1D

and IAH relate to a better sense of control and reduced stress around hypoglycaemia,

positively impacting quality of life. Most patients found CGM and its features helpful to

achieve better glucose control, i.e. ‘being more stable’ and having less or no profound

hypoglycaemia. For this population at risk for recurrent hypoglycaemia the primary

benefits of CGM are feelings of safety and relief rather than convenience. Importantly,

participants reported not only the benefits of CGM for themselves but also their

partners, who are also affected by worries about (nocturnal) hypoglycaemia and the

burden of having to treat the patient in order to restore consciousness.

CGM triggered some patients to engage in understanding trends and more actively

manage their glucose levels, while others reported to respond more passively and

wait for the alarm to go off, before getting into action. Interestingly, these individual

differences in coping with CGM do not predict glycaemic outcomes per se, as all study

161

participants, but one, responded well to real-time CGM in terms of time spent in

normoglycaemia, at least for the study period (Chapter 4). Future studies are needed to

examine how attitudes and coping styles influence self-management behaviour during

as well as after using CGM and how these may enhance improvement of glycaemic

control, both in people with and without IAH.

The improved sense of control many participants (and partners) experienced

concurs with the demonstrated objective glycaemic improvement such as more time

spent in the target range, reduced glycaemic variability, and less frequent nocturnal and

severe hypoglycaemia as shown in Chapter 4. Interestingly, according to the majority

of participants, the improved sense of control washed away after discontinuation of

CGM, implicating there was no psychological legacy effect. Again, this implies that CGM

does not have an effect beyond the actual intervention and that CGM should be used

continuously instead of for a short term.

The combination of the benefit of CGM in terms of improving glycaemic outcomes

(Chapter 4) and the overall positive experience of living with CGM (Chapter 6) strongly

supports the use of CGM in adult patients with T1D and IAH.

Potential barriers for CGM use

To date, relatively few patients with T1D use CGM regularly.26 One major barrier for

continuous use of CGM is the costs associated (i.e., in most countries no reimbursement

is provided by the health insurance companies). In case reimbursement is in place, like

in The Netherlands, only certain patient groups get reimbursement that fulfills strict

indications. Fortunately, “Zorgverzekeraars Nederland” (ZN), the umbrella organization

of nine health insurers in The Netherlands, has very recently approved reimbursement

of CGM for T1D patients with IAH and/or recurring severe hypoglycaemia.

When treating T1D patients with IAH and severe hypoglycaemia in clinical practice,

healthcare professionals frequently first try to improve glycaemia by optimising

self-management (e.g.. by giving structured education regarding flexible insulin

therapy) and changing insulin delivery method from MDI to CSII, before considering

CGM,4 since structured education programs such as HypoAware, DAFNE and BGAT, and

CSII might also prevent severe hypoglycaemia, at least for a proportion of the patients

with severe hypoglycaemia, and cost less than CGM.19,27,28 In Chapter 4 of this thesis, we

showed that in our study population, 18 out of 52 patients (34.6%) used carbohydrate

counting and 23 out of 52 patients (44.2%) were on CSII. Interestingly, we showed equal

benefit of CGM in patients on CSII compared with MDI, and in patients who practised

carbohydrate counting compared with those who did not. This indicates that CGM can

Ch

apte

r 8

162

be used in a wide variety of patients, including those not willing or able to change to

CSII or to practise carbohydrate counting.

It is important to note that CGM can aid patients to self-manage their diabetes

more precisely, provided they are capable of handling the device and data feedback

adequately, with support of a diabetes health care team.29,30 In this context, the question

comes up whether CGM is suitable and beneficial for patients with an unfavourable

psychological profile, e.g. with high psychological distress. Psychological distress

is common in diabetes, including low emotional well-being, high diabetes-related

distress, and fear of hypoglycaemia, negatively affecting patients’ daily self-care and

glycaemic control.31-35 With CGM, patients are faced with real-time feedback on blood

glucose variation and alarms that may be experienced as stressful and difficult to

handle for those with pre-existing high levels of distress.25,29,36 Therefore, in Chapter 5,

we investigated whether psychological distress, operationalized as either low emotional

well-being, high diabetes-related distress, and/or elevated fear of hypoglycaemia

modify the effect of CGM on glycaemic outcomes in patients with T1D and IAH. We

observed lower effect sizes in the low emotional well-being group compared with the

‘normal’ emotional well-being group, which could suggest that CGM is less effective in

patients with low emotional well-being. However, the effect of CGM on the primary

outcome (time in target) remained significant also in patients with low emotional

wellbeing. Furthermore, the lower effect sizes were mainly due to better glycaemic

control during the SMBG phase, rather than to poorer glycaemic control during the

CGM phase. No relevant or consistent interaction effects were observed between

diabetes-related distress or fear of hypoglycaemia and the glycaemic outcomes.

Previous studies have shown that psychological distress is common in diabetes.35

There is little evidence to date on the impact of CGM on distress. Giani and colleagues37

did not find a deterioration of psychosocial functioning in youth with T1D on CGM

for six months. In the DIAMOND study,24 a positive effect of CGM was observed on

diabetes-specific distress but not on other markers of quality of life. To the best of our

knowledge we are the first to study the interaction effect of CGM in high distressed

T1D patients with IAH. As noted by Kubiak et al.38 CGM systems can be experienced as

intrusive causing patients to feel overwhelmed. In this context, research has identified

several themes, such as ‘coping with frustrations’ (confrontation with unexpected

glucose values; false alarms), ‘use of CGM information’ (when to act and how), and

‘technical hassles’ (failures of CGM).25 Our study showed no differences in glycaemic

outcomes between patients with and without psychological distress. Our findings are

reassuring in the sense that the effectiveness of CGM in terms of improved glycaemic

control and less risk of hypoglycaemia is robust, and not attenuated by higher distress

163

at baseline. This would suggest that CGM is ‘safe’ to use in a broad category of type

1 diabetes patients at risk for hypoglycaemia, including those with a less favourable

psychological profile. Whether this is true for more severe psychological or psychiatric

disorders, including major depression and anxiety disorders, is yet unknown. Further

research should help clarify if and to what extent more severe psychological problems

are contra indications for CGM or at least require additional counselling and support.

Implications for clinical practice

How does this contribute to current clinical practice in The Netherlands? The findings

shown in this thesis add further evidence in support of CGM use in patients with T1D

and IAH. It has taken roughly 15 years to provide the final verdict that proper CGM use

truly affects hypoglycaemia, which has been anticipated by the diabetes community

for a long time. Our findings strengthen the evidence in favour of reimbursement of

CGM use in patients with evident impaired awareness of hypoglycaemia. As stated

previously, based on i.a. results of our IN CONTROL trial, regulatory agencies in The

Netherlands have recently approved reimbursement of CGM for T1D patients with IAH

and/or recurring severe hypoglycaemia.

But, does this implicate that clinicians should thus treat all patients with T1D and

IAH with CGM? There are approximately 100.000 patients with T1D in The Netherlands

and roughly 25% of them have IAH. If we were to treat of all 25.000 of them with

CGM continuously (not intermittently or for a short term), this would be a far to

great burden on healthcare cost. One could argue to treat only those patients with

T1D and IAH with CGM who recently have experienced a severe hypoglycaemic event.

During the 16-week SMBG phase in the IN CONTROL trial, 35% of the study population

experienced at least one severe hypoglycaemic event. Extrapolating these results to

real-life, roughly 8750 patients in The Netherlands would qualify for CGM, which is

still a lot from a healthcare cost point of view. We are therefore obligated to search for

other effective, but cheaper options.

Parallel with the IN CONTROL trial, another randomised controlled trial was

performed in the VU University Medical Center (Amsterdam, The Netherlands),

investigating the effectiveness of a brief, partly web-based group intervention

(HypoAware) in a largely comparable population compared to the IN CONTROL trial

(Table 1).19 Interestingly, HypoAware also resulted in fewer severe hypoglycaemic

episodes. In addition, it significantly improved hypoglycaemia awareness, and reduced

hypo-distress in comparison with usual care. We thus now know that, in Dutch patients

with IAH and/or problematic hypoglycaemia, both CGM and HypoAware are effective

interventions. Of course, there are many questions that remain unanswered. Is there

Ch

apte

r 8

164

a difference in effectiveness of CGM versus HypoAware in terms of reducing risk of

severe hypoglycaemia? What intervention is (more) cost-effective? Could we predict,

based on individual patient characteristics, if CGM or HypoAware is more effective in

an individual? And, what is the effectiveness of a combined intervention (structured

education in combination with diabetes technology) in this specific population? Trials

IN CONTROL (n=52) HypoAware (n=121)

Female 24 (46.2) 56 (46.3)

Age, years 48.6 (11.6) 51.3 (13.3)

BMI 25.0 (3.8) 25.4 (4.6)

Type 1 diabetes 52 (100) 121 (100)

Age of diagnosis 16.6 (9.9, 28.0) 24.0 (16.0, 32.5)

Diabetes duration 30.5 (18.5, 40.8) 27.0 (17.0, 35.0)

HbA1c, mmol/mol 58.0 (8.7) 60.4 (11.9)

HbA1c, % 7.5 (0.8) 7.7% (1.1)

Insulin treatment modality

CSII 23 (44.2) 61 (50.4)

MDI 29 (55.8) 60 (49.6)

Self-reported daily home glucose-meter readings, n/day

5.4 (1.9) 4.6 (2.3)

Complications 25 (48.1) 53 (43.8)

Gold score 5.4 (0.7) 5.0 (1.6)

IAH according to Gold 52 (100) 95 (78.5)

Clarke score 5.0 (1.3) 3.7 (1.0)

IAH according to Clarke 45 (86.5) 75 (62.0)

Dutch nationality 48 (92.3) 116 (95.9)

Employed 48 (92.3) 68 (56.2)

Education

No education 0 (0) 4 (3.3)

Lower / vocational 25 (48.1) 30 (24.8)

Higher / pre-university 4 (7.7) 43 (35.5)

University or Professional education 23 (44.2) 44 (36.4)

With partner 40 (76.9) 89 (73.6)

PAID-5 score 6.0 (3.0, 8.8) 6.0 (3.0, 10.0)

PAID-5, distressed 17 (32.7) 46 (38.0)

HFS-W score* 36.5 (25.5, 51.0) 36.1 (20.1, 49.3)

Data are n (%), mean (SD) or median (25%, 75%). BMI: body mass index; CSII: continuous subcutaneous insulin infusion;

MDI: multiple daily injection; IAH: impaired awareness of hypoglycaemia; PAID-5: Problem Areas in Diabetes 5 items

questionnaire; HFS-W: Hypoglycaemia Fear Survey, Worry subscale, *transformed to a 0 to 100 scale.

Table 1. Socio-demographic and clinical characteristics of the IN CONTROL and HypoAware population

165

providing answers to these question should be prioritised.

We propose a stepped-care approach for clinicians when treating adult patients with

T1D and IAH. First, we would like to emphasise the importance of frequent and close

patient-provider contact. Although CGM use was suboptimal in the HypoCOMPaSS

study, Little and colleagues were able to show a tremendous drop in the rate of

severe hypoglycaemia using strategies available in clinical practice, underscoring the

importance of close and frequent contact.10 Close patient-provider contact should

therefore always be the main component of diabetes care. Since it seems common

sense that a brief group intervention as HypoAware costs less than CGM, certainly on

the long-term, we propose that when faced with a patient with IAH and/or problematic

hypoglycaemia despite close contact, clinicians should consider starting with a

structured education program such as HypoAware. The problem of hypoglycaemia

(awareness) will probably be solved in a proportion of those patients, but certainly not

in all. Therefore, continuous glucose monitoring should be available for all patients

with T1D in whom hypoglycaemia (awareness) remains a major and devastating

problem.

Ch

apte

r 8

166

1. Lawrence RD. Insulin

hypoglycaemia: changes in nervous

manifestations. Lancet 1941; (2): 602.

2. McCoy RG, Van Houten HK,

Ziegenfuss JY, Shah ND, Wermers

RA, Smith SA. Increased mortality

of patients with diabetes reporting

severe hypoglycemia. Diabetes Care

2012; 35(9): 1897-901.

3. Cryer PE. Severe hypoglycemia

predicts mortality in diabetes.

Diabetes Care 2012; 35(9): 1814-6.







38(8): 1592-609.

5. Pickup JC, Sutton AJ. Severe

hypoglycaemia and glycaemic

control in Type 1 diabetes:

meta-analysis of multiple daily

insulin injections compared with

continuous subcutaneous insulin

infusion. Diabet Med 2008; 25(7):

765-74.






CD008101.







36(12): 4160-2.









1240-7.









Diabet Med 2010; 27(6): 666-72.











REFERENCES

167

37(8): 2114-22.




7(5): 493-503.





Med 2010; 363(4): 311-20.






2006; 29(12): 2730-2.




diabetes. N Engl J Med 2008; 359(14):

1464-76.

15. Kilpatrick ES, Rigby AS, Goode K,

Atkin SL. Relating mean blood

glucose and glucose variability to

the risk of multiple episodes of

hypoglycaemia in type 1 diabetes.

Diabetologia 2007; 50(12): 2553-61.

16. Cox DJ, Kovatchev BP, Julian

DM, et al. Frequency of severe

hypoglycemia in insulin-dependent

diabetes mellitus can be predicted

from self-monitoring blood glucose

data. J Clin Endocrinol Metab 1994;

79(6): 1659-62.

17. Qu Y, Jacober SJ, Zhang Q, Wolka LL,

DeVries JH. Rate of hypoglycemia in

insulin-treated patients with type

2 diabetes can be predicted from

glycemic variability data. Diabetes

Technol Ther 2012; 14(11): 1008-12.






Diabetologia 2012; 55(12): 3155-62.

19. Rondags SM, de Wit M, Twisk

JW, Snoek FJ. Effectiveness

of HypoAware, a Brief Partly

Web-Based Psychoeducational

Intervention for Adults With Type 1

and Insulin-Treated Type 2 Diabetes



Diabetes Care 2016; 39(12):90-96








50-2.

21. Cranston I, Lomas J, Maran A,

Macdonald I, Amiel SA. Restoration

of hypoglycaemia awareness in

patients with long-duration insulin-

dependent diabetes. Lancet 1994;

344(8918): 283-7.






Ch

apte

r 8

168


Diabetes Care 2010; 33(10): 2175-7.







28(10): 1158-67.

24. 24. Polonsky WH, Hessler D,

Ruedy KJ, Beck RW. The Impact of

Continuous Glucose Monitoring on

Markers of Quality of Life in Adults

With Type 1 Diabetes: Further

Findings From the DIAMOND

Randomized Clinical Trial. Diabetes

Care 2017; 40(6): 736-41.






2015; 38(4): 544-50.






971-8.





Diabetes Care 2001; 24(4): 637-42.

28. Shearer A, Bagust A, Sanderson D,

Heller S, Roberts S. Cost-effectiveness

of flexible intensive insulin

management to enable dietary

freedom in people with Type 1

diabetes in the UK. Diabet Med 2004;

21(5): 460-7.





2010; 27(9): 1060-5.

30. Peters AL, Ahmann AJ, Battelino T, et

al. Diabetes Technology-Continuous

Subcutaneous Insulin Infusion

Therapy and Continuous Glucose

Monitoring in Adults: An Endocrine

Society Clinical Practice Guideline. J

Clin Endocrinol Metab 2016; 101(11):

3922-37.



with diabetes. J Clin Psychol 2001;

57(4): 457-78.


KE, de GM, Carney RM, Clouse RE.

Depression and poor glycemic

control: a meta-analytic review of

the literature. Diabetes Care 2000;

23(7): 934-42.

33. Wild D, von MR, Brohan E,

Christensen T, Clauson P, Gonder-

Frederick L. A critical review of the

literature on fear of hypoglycemia in



Patient Educ Couns 2007; 68(1): 10-5.

34. Barendse S, Singh H, Frier

BM, Speight J. The impact of

hypoglycaemia on quality of life and

169

related patient-reported outcomes in

Type 2 diabetes: a narrative review.

Diabet Med 2012; 29(3): 293-302.

35. Fisher L, Hessler D, Polonsky W,

Strycker L, Masharani U, Peters A.




Complications 2016; 30(6): 1123-8.


KM, Naranjo D, Bensen R, Hood KK.

Diabetes Device Use in Adults With

Type 1 Diabetes: Barriers to Uptake

and Potential Intervention Targets.

Diabetes Care 2017; 40(2): 181-7.

37. Giani E, Snelgrove R, Volkening

LK, Laffel LM. Continuous Glucose

Monitoring (CGM) Adherence

in Youth With Type 1 Diabetes:

Associations With Biomedical and

Psychosocial Variables. J Diabetes Sci

Technol 2016.

38. Kubiak T, Mann CG, Barnard KC,

Heinemann L. Psychosocial Aspects

of Continuous Glucose Monitoring:

Connecting to the Patients'

Experience. J Diabetes Sci Technol

2016; 10(4): 859-63.

Ch

apte

r 8

170

Nederlandse samenvatting

Summary in Dutch

Dankwoord

Acknowledgements

Author affiliations

Biografie

Biography

172

Ned

erla

nd

se s

amen

vatt

ing

NEDERLANDSE SAMENVATTING

Type 1 diabetes

Het eten van koolhydraten (bijvoorbeeld aardappelen, brood, pasta of rijst) zorgt

voor een stijging van glucose in het bloed (bloedsuiker). Onze lichaamscellen, met name

onze hersenen, hebben glucose hard nodig omdat glucose de belangrijkste brandstof is

voor onze lichaamscellen. Het is daarom van levensbelang dat glucose in onze lichaams-

cellen terecht kan komen. Insuline, een hormoon dat de alvleesklier aanmaakt, zorgt

ervoor dat 1) glucose in de lichaamscellen opgenomen kan worden waardoor 2) de

glucosewaarde in het bloed daalt. Zonder insuline stijgt de glucosewaarde dus in het

bloed en hebben de lichaamscellen geen brandstof waardoor een lichaam verzuurt.

Zonder insuline kan de mens niet leven.

Diabetes Mellitus, of suikerziekte, is een ziekte die wereldwijd, en in Nederland,

steeds vaker voorkomt. Bij mensen met suikerzieke is de glucosewaarde in het bloed

te hoog omdat het lichaam ongevoelig is geworden voor insuline (type 2 diabetes) of

omdat het lichaam helemaal geen insuline meer aanmaakt (type 1 diabetes [T1D]).

Omdat je zonder insuline niet kan leven is het van levensbelang dat mensen met T1D

hun hele leven zelf insuline toedienen. Dit kan door middel van meerdere injecties met

kort- en langwerkende insuline per dag of via continue toediening van insuline via een

insulinepomp.

De glucosewaarde in het bloed schommelt continu en er zijn erg veel factoren van

invloed op de glucosewaarde. Sommige factoren zorgen ervoor dat glucose in het

bloed stijgt (o.a. eten, stress, ziekte, bepaalde medicijnen) en andere factoren zorgen

er weer voor dat glucose in het bloed daalt (o.a. insuline, alcohol, sport). Bij mensen

zonder diabetes wordt de glucosewaarde, ondanks al deze factoren, heel nauwkeurig

binnen normaalwaarden gehouden. Bij mensen met T1D wordt met insulinetherapie

geprobeerd deze ‘normale situatie’ zo goed mogelijk na te bootsen. Helaas blijkt dit

vandaag de dag nog steeds onhaalbaar. Mensen met T1D moeten namelijk iedere dag,

op meerdere momenten, zelf bepalen hoeveel insuline ze zichzelf toe moeten dienen,

rekening houdend met al die factoren die van invloed zijn op de glucosewaarde.

De hoeveelheid insuline die men toe moet dienen is namelijk afhankelijk van al

die bekende (en onbekende) factoren die van invloed zijn op de glucosewaarde,

waaronder: de hoeveelheid suiker in het eten, de bloedglucosewaarde voor het eten,

wel of niet van plan zijn om te gaan sporten, nachtdiensten, ander medicijngebruik,

het uitlaten van de hond, stress op het werk, wind mee of wind tegen, werken in de

tuin, het drinken van een borreltje na het werk, dansen op een festival en nog vele

anderen. Het is daarom onvermijdelijk dat, ondanks het feit dat mensen met T1D met

173

Sum

mar

y in

Du

tch

zó veel zaken rekening houden, de glucosewaarde vaak naar boven (hyperglykemie) of

naar beneden (hypoglykemie) doorschiet. Zowel hypoglykemieën als hyperglykemieën

zijn potentieel levensgevaarlijk.

Hyperglykemieën geven veel klachten, zoals veel plassen, veel dorst, vermoeidheid

en misselijkheid. Onderzoek heeft tevens onomstotelijk aangetoond een hoog ‘lange

termijn suikergehalte’ (HbA1c; de gemiddelde suikerwaarde van de afgelopen

3 maanden) het risico op complicaties van suikerziekte (blindheid, nierziekten,

zenuwproblemen en ziekten van de grote en kleine bloedvaten) sterk verhoogd.

Het is daarom erg belangrijk dat mensen met T1D zo goed mogelijk voorkomen

dat hun glucosewaarde naar boven doorschiet. Het nadeel van het voorkomen van

doorschieters naar boven zijn echter doorschieters naar beneden (hypoglykemieën).

Hypoglykemieën zijn de grote keerzijde van insulinetherapie.

Hypoglykemieën en ‘hypo-unawareness’

Een hypoglykemie is een te lage glucosewaarde. Dit is potentieel levensgevaarlijk

omdat het lichaam (met name de hersenen) niet kan functioneren zonder glucose.

Hypoglykemieën geven veel klachten, zoals zweten, trillen, vermoeidheid, honger,

hoofdpijn, en plotselinge wisselingen van het humeur en hebben dus een grote

invloed op het dagelijks leven. Bij een hypoglykemie kan het denkvermogen ook

sterk verslechteren, wat bijvoorbeeld tot ongelukken kan leiden (denk bijvoorbeeld

aan een hypoglykemie in het verkeer). Als de glucosewaarde nog verder zakt kan

iemand in coma raken of (heel soms) zelfs overlijden! De klachten die hypoglykemieën

veroorzaken zijn erg vervelend, maar ook erg belangrijk. Deze klachten (symptomen)

kunnen mensen namelijk alarmeren over het feit dat hun bloedsuikerwaarde te laag

is waardoor ze de hypoglykemie zelf kunnen behandelen (door het eten/drinken van

koolhydraten [suiker]).

Maar liefst een kwart van de volwassenen met T1D voelt hypoglykemieën echter

niet (meer) goed aankomen. Dit verminderde bewustzijn van hypoglykemieën wordt

‘hypo-unawareness’ genoemd en komt vaak voor bij mensen die langdurig suikerziekte

hebben. Door de vele hypoglykemieën door de jaren heen is het lichaam als het ware

gewend geraakt aan hypoglykemieën, waardoor het lichaam minder hypo-symptomen

maakt (zoals zweten, trillen etc.). Als iemand geen hypo-symptomen krijgt, wordt

hij/zij dus niet gealarmeerd over het lage bloedsuiker. Dit kan ervoor zorgen dat de

glucosewaarde in het bloed dusdanig laag worden dat iemand buiten bewustzijn

raakt. Uit eerder onderzoek blijkt dan ook dat mensen met hypo-unawareness een

sterk verhoogd risico hebben op ernstige hypoglykemieën. Ernstige hypoglykemieën

zijn hypoglykemieën waarbij mensen hulp van anderen (bijvoorbeeld ambulanceper-

174

Ned

erla

nd

se s

amen

vatt

ing

soneel, partners, collega’s) nodig hebben om de hypoglykemie te behandelen.

Continue glucose monitoring

Omdat de glucosewaarde continu schommelt moeten mensen met T1D meerdere

keren per dag hun glucosewaarde in de gaten houden. Dit doen de meeste mensen

met vingerprikken. Door met een dun naaldje in een vinger te prikken komt er een

druppeltje bloed vrij. Een glucosemeter meet vervolgens de glucosewaarde uit dat

druppeltje bloed. Met deze vingerprikken kunnen mensen dus ongeveer vier keer per

dag een glucosewaarde zien. Deze vingerprikken zijn echter vervelend, pijnlijk en alle

schommelingen tussen die metingen door worden gemist met vingerprikken. Het is

alsof mensen de hele dag auto (moeten) rijden, niet te hard, maar ook niet te zacht,

maar in plaats van continu te zien hoe hard ze gaan kunnen ze maar 4 keer per dag

hun snelheid controleren.

Sinds een jaar of 10 zijn er continue glucose monitoren (CGM) beschikbaar. Deze

systemen bestaan uit 1) een glucosesensor (meet glucosewaardes continu in het

vet onder de huid), 2) een zender (stuurt de glucosewaarde naar de monitor) en 3)

een monitor. Op de monitor is de actuele glucosewaarde te zien, een grafiek, en een

trendpijl waardoor mensen kunnen zien of hun glucose stabiel is, stijgt, of daalt. Een

zeer belangrijk voordeel van CGM t.o.v. vingerprikken is het feit dat CGM mensen kan

waarschuwen (met alarmen) dat hun glucosewaarde te laag of te hoog is of wordt. Dit

kan bijvoorbeeld waardevol zijn in de nacht, wanneer mensen vaak langdurig hun

glucosewaarde niet controleren. CGM is hiermee in potentie dus een zeer waardevol

hulpmiddel voor de behandeling van type 1 diabetes.

Onderzoeken hebben consequent aangetoond dat CGM het lange termijn suiker

(HbA1c) verlaagt zonder dat CGM het aantal hypoglykemieën verhoogt. Er werd ook

verwacht dat CGM het aantal ernstige hypoglykemieën zou verlagen, maar dit werd

in al die onderzoek niet aangetoond. Bij veel onderzoeken naar het effect van CGM

werden mensen met 1) een recente ernstige hypoglykemie of 2) hypo-unawareness

echter uitgesloten van deelname. Dit zorgde ervoor dat er überhaupt zeer weinig

ernstige hypoglykemieën voorkwamen in al die onderzoeken. En inderdaad, je kunt

geen vermindering aantonen op iets dat niet voorkomt.

Doel van het proefschrift

Het doel van het proefschrift was om het effect van CGM te onderzoeken op de

glykemische instelling (een combinatie van factoren dat samen een indicatie geeft

over hoe goed de diabetes ingesteld is) en kwaliteit van leven bij mensen met T1D en

hypo-unawareness.

175

Sum

mar

y in

Du

tch

In Hoofdstuk 2 van dit proefschrift geven we een overzicht van de bestaande

onderzoeken die al gedaan zijn naar het effect van CGM bij mensen met T1D. Daarbij

letten we met name op de hypoglykemie-gerelateerde uitkomsten van de onderzoeken.

We zien dat de meeste onderzoeken geen vermindering van het aantal (ernstige)

hypoglykemieën laten zien. Echter, de paar studies die focusten op het verminderen

van hypoglykemieën tonen dit wél aan. Eén onderzoek dat het effect van CGM met een

laag-glucose stop functie (een functie waarbij de toediening van insuline automatisch

stopt als een glucosewaarde te laag wordt) onderzocht liet een vermindering van het

aantal ernstige hypoglykemieën zien bij jonge kinderen. We concludeerden daarom

dat CGM alleen effectief is in het verminderen van hypoglykemieën als mensen hem

specifiek voor dit doel gebruiken. Verder concludeerden we dat het onbekend bleef

of CGM ernstige hypoglykemieën vermindert bij volwassenen met hypo-unawareness.

Het IN CONTROL onderzoek

In Hoofdstuk 3 presenteren we de achtergrond en het ontwerp van het IN CONTROL

onderzoek. Zoals gezegd, hypo-unawareness komt ongeveer bij een kwart van de

mensen met type 1 diabetes voor. Dit zorgt voor een sterk verhoogd risico op ernstige

hypoglykemieën. Het effect van CGM op glykemische instelling, kwaliteit van leven en

hypo-awareness is onbekend bij deze hoog-risico groep. Om dit effect te onderzoeken

hebben we het IN CONTROL onderzoek opgezet. Er deden 52 volwassen mensen met

T1D en hypo-unawareness mee aan het onderzoek. Het IN CONTROL onderzoek was

een gerandomiseerde, cross-over studie met twee onderzoeksperiodes van 16 weken.

Tijdens de ene onderzoeksperiode controleerden de deelnemers hun glucosewaarde

met standaard vingerprikken en tijdens de andere periode met CGM. Een loting

bepaalde de volgorde van de onderzoeksperiodes. Tussen de onderzoeksperiodes

zat een periode van 12 weken waarbij geen onderzoeksactiviteiten plaatsvonden

(uitwasperiode). Tijdens beide periodes werd de glykemische instelling (o.a. de

hoeveelheid tijd per dag dat de glucosewaarde binnen, onder en boven de normaal-

waarden was, en glucose variabiliteit), en het aantal ernstige hypoglykemieën gemeten.

Tevens werd de kwaliteit van leven en hypo-awareness gemeten voor en na beide

onderzoeksperiodes. Tijdens beide onderzoeksperiodes bezochten de deelnemers het

onderzoekscentrum maandelijks.

Het effect van CGM

Vervolgens bespreken we in Hoofdstuk 4 de resultaten van het IN CONTROL

onderzoek. We vonden dat CGM, vergeleken met standaard vingerprikken, ervoor

zorgt dat de glucosewaarde van mensen met T1D en hypo-unawareness meer tijd

per dag binnen de normaalwaarden zit, door vermindering van de tijd onder én een

176

Ned

erla

nd

se s

amen

vatt

ing

vermindering van de tijd boven de normaalwaarden. Een zeer belangrijke bevinding

was het feit dat er met CGM minder ernstige hypoglykemieën voorkwamen én dat met

CGM minder mensen ernstige hypoglykemieën ervaren. Tevens verminderde CGM het

aantal nachtelijke hypoglykemieën, de diepte en de duur van hypoglykemieën, en de

glucose variabiliteit (schommelingen van glucose). We zagen dat CGM net zo effectief

is bij mensen die worden behandelend met een insulinepomp vergeleken met mensen

die behandeld worden met insuline injecties. Helaas zagen we ook dat het effect van

CGM direct verdwijnt als mensen weer teruggaan naar standaard vingerprikken.

Het is dus niet zinvol om CGM een korte periode te gebruiken. We concludeerden dat

CGM een zeer waardevol hulpmiddel is voor behandeling van mensen met T1D en

hypo-unawareness.

In Hoofdstuk 4 en Hoofdstuk 5 bespreken we tevens het effect van CGM op kwaliteit

van leven. CGM zorgt ervoor dat mensen met T1D en hypo-unawareness zich minder

zorgen maken over hypoglykemieën (minder angst hebben voor hypoglykemieën).

Verrassend genoeg zagen we dat CGM geen relevant effect heeft op andere maten

van kwaliteit van leven (vragenlijsten), zoals diabetes gerelateerde stress, emotioneel

welzijn of vertrouwen in diabetes zelfzorg. Mogelijk komt dit omdat het kwaliteit van

leven voor de start van het onderzoek al hoog was.

Ervaringen van patiënten

CGM is niet een passieve interventie, zoals bijvoorbeeld tabletten dat zijn, maar het

maakt een belangrijk deel uit van het dagelijks leven, zowel negatief als positief. CGM

doet zelf namelijk niets met de glucosewaarde. Het meet alleen glucose, maar past de

dosering insuline zelf niet aan. Veranderingen in zelfzorggedrag, ingegeven door CGM,

moeten voor verbetering van de glykemische instelling zorgen. Het is daarom van zeer

groot belang om te begrijpen hoe mensen het gebruik van CGM ervaren. Als mensen

CGM bijvoorbeeld maar vervelend vinden, omdat het continu alarmen afgeeft, en CGM

wordt maar half gebruikt, zal het ook niet voor verbetering van de instelling zorgen.

In Hoofdstuk 6 hebben we daarom de verwachtingen van CGM en ervaringen met

CGM onderzocht bij de deelnemers van het IN CONTROL onderzoek d.m.v. interviews.

Drieëntwintig deelnemers van het IN CONTROL onderzoek werden geïnterviewd.

Dit interview ging over de verwachtingen van CGM en de ervaringen met CGM. We

concludeerden dat de voordelen van CGM m.b.t. het verbeteren van de glykemische

instelling en het voorkómen van ernstige hypoglykemieën duidelijk gepaard gaat met

een beter ‘gevoel van controle’ met CGM. Een beter gevoel van controle komt voor een

belangrijk deel door het feit dat glucosewaardes continu te zien zijn op de monitor en

het feit dat mensen wisten dat ze gealarmeerd worden als hun glucosewaarde de bocht

uit zou vliegen. De deelnemers meldden meerdere aspecten van een beter ‘gevoel van

177

Sum

mar

y in

Du

tch

controle’, zoals: ‘stabieler zijn’ (minder schommelingen), minder (nachtelijke) hypogly-

kemieën hebben, en het verkrijgen van meer inzicht in hun diabetes (bijv. inzicht in

dagelijkse patronen en patronen rondom sport en alcoholgebruik). De deelnemers

ervaarden daarnaast een positieve invloed van CGM op hun psychologische welzijn.

Dit kwam omdat ze minder gestrest en angstig waren en ze zich veiliger, vrijer en

onafhankelijker voelden. Ze ervaarden tevens dat het welzijn van hun partners en

kinderen verbeterden. Uiteraard waren er ook negatieve ervaringen met CGM. Dit

waren met name technische problemen en de vele alarmen. Echter, bijna iedereen

vond de positieve aspecten een stuk zwaarder wegen dan de negatieve. Alle deelnemers

zeiden dat ze CGM zeker zouden adviseren aan andere mensen met T1D en hypo-una-

wareness.

Mogelijke barrières voor het gebruik van CGM

Vandaag de dag gebruiken relatief weinig mensen CGM. Een belangrijke barrière

voor het gebruik van CGM was het feit dat CGM niet vergoed werd voor mensen met

T1D en hypo-unawareness. Veel mensen kunnen CGM niet zelf betalen. Gelukkig

hebben de verzekeraars ‘hypo-unawareness’ recent toegevoegd als vergoedingscri-

terium, waardoor CGM nu vergoed wordt voor deze mensen, hetgeen waarschijnlijk

tot meer gebruik van CGM zal leiden.

Het hebben van diabetes gaat relatief vaak gepaard met psychologische problemen,

zoals angst voor hypoglykemieën of complicaties, of een sombere stemming

(emotioneel welzijn). We weten dat deze problemen de diabetes zelfzorg negatief

kunnen beïnvloeden. Een sombere stemming zou er bijvoorbeeld voor kunnen zorgen

dat iemand CGM minder vaak gebruikt, waardoor de effectiviteit van CGM minder goed

is dan bij mensen met een opgewekte stemming. In Hoofdstuk 5 hebben we daarom

onderzocht of een sombere stemming, diabetes-gerelateerde stress, of angst voor

hypoglykemieën invloed hebben op de effectiviteit van CGM. We zagen dat het effect

van CGM iets minder groot was bij mensen met een sombere stemming, vergeleken met

mensen met normaal emotioneel welzijn. Het effect van CGM bij mensen met sombere

stemming was echter niet volledig verdwenen. Daarnaast zagen we dat het effect van

CGM was even groot bij mensen met veel en weinig diabetes-gerelateerde stress en bij

mensen met veel en weinig angst voor hypoglykemieën. We concludeerden daarom

dat CGM prima ingezet kan worden bij mensen met een psychologisch kwetsbaarder

profiel. Het hebben van een kwetsbaar psychologisch profiel mag dus geen barrière

zijn voor artsen om een patiënt met T1D en hypo-unawareness te behandelen met

CGM.

178

Ned

erla

nd

se s

amen

vatt

ing

Conclusie

In dit proefschrift hebben we het effect van CGM onderzocht op glykemische

instelling en kwaliteit van leven bij mensen met T1D en hypo-unawareness. Allereerst

toonden we zeer consistent aan dat CGM, vergeleken met standaard vingerprikken, alle

glykemische uitkomstmaten verbetert. In het bijzonder zagen we een vermindering

van het aantal ernstige hypoglykemieën met CGM. Dit zorgde er ook voor dat mensen

zich met CGM minder zorgen maken met CGM. Tevens lieten we zien dat CGM even

effectief is bij mensen die behandeld worden met een insulinepomp als bij mensen die

behandeld worden met insuline injecties. Helaas zagen we ook dat het effect van CGM

volledig verdwijnt als je CGM weer afpakt. Verder hebben we aangetoond dat CGM

effectief is voor mensen met en zonder een kwetsbaar psychologisch profiel. Een zeer

belangrijke bevinding is het feit dat veruit de meeste patiënten CGM als zeer positief

ervaren. CGM is dus een zeer waardevol hulpmiddel voor de behandeling van mensen

met type 1 diabetes met hypo-unawareness.

179

Sum

mar

y in

Du

tch

180

Au

thor

Aff

ilia

tion

s

AUTHOR AFFILIATIONS

Cornelis A.J. van Beers

Department of Internal Medicine, VU University Medical Center

Amsterdam, The Netherlands

Wilmy Cleijne

Department of Medical Psychology, VU University Medical Center


J. Hans DeVries

Department of Endocrinology, Academic Medical Center, University of Amsterdam,


Michaela Diamant

Diabetes Center / Department of Internal Medicine, VU University Medical Center\


Petronella H. Geelhoed-Duijvestijn

Department of Internal Medicine, Medical Center Haaglanden

The Hague, The Netherlands

Susanne J. Kleijer



Mark H.H. Kramer



Stefanie M. Rondags



Erik H. Serné



181

Au

thor

Aff

ilia

tion

s

Mark M. Smits



Frank J. Snoek

Department of Medical Psychology, VU University Medical Center, Amsterdam, The

Netherlands; Department of Medical Psychology, Academic Medical Center, University

of Amsterdam


Anne Vloemans



Maartje de Wit



182

Dan

kwoo

rd

DANKWOORD

Save the best for last…

Allereerst wil ik alle deelnemers aan het IN CONTROL onderzoek bedanken voor het

deelnemen én volhouden van het onderzoek. Wij hebben nogal wat van jullie gevraagd.

Veertien keer naar het onderzoekscentrum komen, 13 telefoontjes, wekelijks glucose-

gegevens uploaden waarna soms/vaak bleek dat al die gegevens verdwenen waren...

zucht. Het was af en toe een hele opgave. Bedankt dat ik zoveel van jullie heb mogen

leren. Beste Margreet, bedankt dat je wilde poseren voor de kaft van het proefschrift.

Ik vond het erg prettig om met je samen te werken.

Ook wil ik alle leden van de leescommissie bedanken voor het lezen en beoordelen

van dit proefschrift.

En dan natuurlijk mijn promotieteam. Beste Michaela, ik heb je helaas niet lang mogen

kennen en daarom helaas te weinig van je kunnen leren. Bedankt voor de kans die je

me gegeven hebt. Ik ben stiekem best trots dat ik één van de laatste ben die onderdeel

uitmaakt van de ‘Michaela-groep’. Beste Mark, bedankt dat je er voor me was wanneer

ik het nodig had. Bijvoorbeeld helpen met contractverlengingen, mij introduceren

in het Horacio E. Oduber Hospitaal op Aruba, de mogelijkheid om vóór mijn buiten-

landavontuur te solliciteren voor de opleiding. Voor mij allemaal belangrijke zaken

waarbij het erg fijn is dat je me geholpen hebt. Bedankt. Beste Frank, na het overlijden

van Michaela twijfelde ik af en toe of het wel goed ging komen met het onderzoek. Ik

weet nog goed dat ik weg liep van ons kennismakingsgesprek met een glimlach op mijn

gezicht en een rechte rug, omdat jij mij de rust en het vertrouwen gaf dat het zeker

goed ging komen. En zo is dat eigenlijk altijd gebleven. Na iedere projectbespreking

ging ik weer met goede moed aan de slag. Heel erg bedankt daarvoor. Beste Erik,

bedankt dat ik altijd en voor alles je kamer binnen kon stormen, ondanks jouw volle

agenda. Bedankt voor je scherpe analyses van de data en voor je hulp om de stukken

naar een hoger niveau te tillen. Daarnaast uiteraard bedankt voor de gezelligheid

tijdens ski-reisjes en congressen. Beste Nel, bedankt voor je vertrouwen in mij en dat je

mij voorgedragen hebt bij Michaela toen er een promotieplek beschikbaar kwam. Een

klinisch onderzoek waarvan jij al dacht dat het goed bij me zou passen. Bedankt voor al

je hulp mij beter te maken in het begeleiden van mensen met type 1 diabetes en voor je

hulp om meer mensen in het onderzoek te krijgen. Daarnaast natuurlijk bedankt voor

je hulp bij mijn sollicitatie voor de opleiding. Beste Hans, officieus behoor je natuurlijk

183

Ack

now

led

gem

ents

ook bij het promotieteam. Bedankt dat ik bij jou aan heb mogen kloppen voor hulp.

Ik vond het erg prettig om stukken met je te schrijven, omdat je ontzettend scherp

en snel reageert. Ik ben ervan overtuigd dat mede door dit schrijfproces met jou het

publiceren van de manuscripten een stuk makkelijker is gegaan. Bedankt.

Ook wil ik alle internisten en diabetesverpleegkundigen van het VUmc en HMC

Westeinde bedanken voor hun inzet om mensen aan te melden voor het onderzoek.

Beste Judith en Jolanda, in het begin wist ik niet waar ik moest beginnen met al die

sensor-uitslagen. Jullie hebben mij geleerd om hier structuur in aan te brengen, wat

het begeleiden van de deelnemers een stuk makkelijker werd. Beste Richard, allereerst

natuurlijk bedankt voor het goedkeuren van dit proefschrift. Ook bedankt voor alle

korte overlegmomentjes stiekem tussen neus en lippen door. Deze hebben me door

de jaren heen wel degelijk verder geholpen. Bedankt voor de gezelligheid tijdens

congressen. Ik denk nog altijd met plezier terug aan San Fransisco.

Beste Lilian, jij hebt ons allemaal enorm geholpen door de jaren heen. Bedankt dat

je over ons allemaal gewaakt hebt en dat je er bovenop zat als er iets dreigde mis te

gaan. Geniet van je pensioen, je zal gemist worden op het VUmc.

Ook wil ik alle research assistenten bedanken voor hun hulp bij het onderzoek.

Zonder jullie hulp is het leven van een arts-onderzoeker echt een stuk minder

aangenaam.

Daarnaast wil ik graag alle onderzoekers van de medische psychologie bedanken

voor hun hulp bij het opzetten van het kwalitatieve onderzoek. Beste Stefanie, bedankt

voor het afnemen van alle interviews. Dit was nog best een klus. Beste Maartje en

Anne, onwijs bedankt dat jullie mij zo geholpen hebben in de periode dat ik op Aruba

zat. Het was niet makkelijk om in een ander land met een nieuw baan bezig te zijn én

de laatste stukken af te maken. Ik denk dat het tot mooie resultaten heeft geleid.

En dan natuurlijk mijn eigen collega’s. Susan, Nalini en Jolanda, bedankt dat jullie

mij zo goed op weg hebben geholpen en dat ik altijd bij jullie aan mocht kloppen voor

vragen over deelnemers, het protocol etc. Ik heb in korte tijd veel van jullie mogen

leren. Beste Susan, ik ben blij dat je nu zo goed op je plek bent. Annelies, ik vond je altijd

een fijne, rustige collega. Het feit dat we beide in Den Haag woonden schiep natuurlijk

wel een band. En wat knap van je dat je het met mij hebt uitgehouden op maar liefst

twee kamers. Dat geldt natuurlijk ook voor Liselotte. Bedankt voor alle snapchats en

184

Dan

kwoo

rd

vunzige grapjes. Jij kan zo meekomen in iedere willekeurige voetbalkantine. Maar,

zonder grap, ook serieus bedankt voor je steun, op werkgebied en daarbuiten. Anna,

bedankt dat je een kartrekker bent als het om borrelen en gezelligheid gaat...want dat

konden we bij het VUmc vaak wel gebruiken...(inderdaad, dit is een zuur grapje). Erik

van B!, thanks voor het luisteren naar mijn geouwehoer en kerstmuziek, en naar mijn

weerstand bij, naja, allerlei zaken. Fijn dat je door de weerstand heen wist te ploeteren

als ik je dan later tóch vroeg me te helpen om plaatjes, zinnen, grafieken etc. mooier

te maken. Bedankt dat je zo’n fijne kamergenoot was. En dan die Smitsie, collegialer

worden ze niet gemaakt. Je bent nooit te beroerd om wie dan ook verder te helpen,

zelfs als het je een hele zondag kost. Én dit doe je zonder dat je er standaard iets voor

terug vraagt. Daarom vind ik het zo leuk dat we nu samen op het mooiste stuk van dit

proefschrift staan. Lennart en Gerard, jullie werkten samen vaak op mijn lachspieren.

Bedankt dat je zo fijn publiek bent voor mijn grapjes. Ik leef op goed publiek. Bedankt

voor de gesprekken die ook níet over het onderzoek gingen. Beste Marcel, bedankt

voor het sparren, voor je feedback, en bedankt dat je me naar het AMC trapte om

eens bij Hans de Vries aan te kloppen. Beste Nick, bedankt voor alle gezellige feestjes,

etentjes en lunches. Daarnaast ontzettend bedankt voor je opvang in het OLVG West

in de eerste maanden van mijn opleiding en tijdens mijn eerste diensten. Beste Jonne,

ook jij bent nooit te beroerd om mensen verder te helpen, zonder eigen agenda. Je bent

de vriendelijkheid zelve en inhoudelijk super sterk. Bedankt dat je zo’n fijne collega

was. Lieve Martine, bedankt dat je een goede vriend van me bent geworden. Jij was

één van de mensen die het minder eng maakte om naar Amsterdam te verhuizen.

(Want Martine woont er tenminste ook.) Ik vind het heel bijzonder dat ik de hele

dag op je bruiloft was, dit maakte me stiekem een beetje trots. Én het is natuurlijk

hilarisch dat we samen in dit boekje staan. (Hoofdstuk 7 lieve lezers, maar dit wisten

jullie natuurlijk al, want niemand begint een boek achterin.) Bedankt voor een te gek

weekend op Tenerife en bedankt dat ik altijd bij je aan kan kloppen, voor leuke en

minder leuke dingen, zowel professioneel als persoonlijk.

Ik was nooit dit pad ingeslagen als ik het niet zo naar mijn zin had gehad als

assistent interne geneeskunde in het Haaglanden Medisch Centrum. Daarom wil in

alle internisten van het HMC bedanken door mij te laten zien wat een mooi vak de

interne geneeskunde is. En natuurlijk mijn véél te gezellige collega’s van het HMC. Ik

heb nog steeds hoofdpijn van de Erdingers in de Boterwaag en de Tequilla shots in

the Millers. Lieve Freek, Frank, Anne Floor en Josien, fijn dat we ondanks alle drukke

agenda’s elkaar niet uit het oog verliezen. Het is moeilijk om vier arts-assistenten op

één plek op één dag en op hetzelfde tijdstip bij elkaar te krijgen, maar het lukt af en

185

Ack

now

led

gem

ents

toe toch. En dat is een feestje. Rutger, buddy van het eerste uur, bedankt voor al je

hulp in de eerste periode van mijn assistentschap en het vertrouwen dat je altijd in

me hebt gehad. Bedankt voor alle gezelligheid en natuurlijk bedankt dat ik bij je kon

onderduiken toen dat even nodig was.

Ook wil ik alle collega’s van het Horacio E. Oduber Hospitaal, Aruba, enorm bedanken

voor een zeer bijzondere, chaotische, leerzame en gezellige ervaring! Ik hoop over een

aantal jaren zo af en toe eens terug te komen. Dit geldt ook voor alle nieuwe collega’s

van het OLVG West. Bedankt voor jullie warme ontvangst en een fijn begin van mijn

opleiding. Beste Arne Jon, echt top dat onze paden elkaar weer kruisen na het Haagsche

avontuurtje. Ik kijk er naar uit de komende jaren lekker de nerd met je uit te hangen en

veel van je te kunnen leren. Jorrit, studiemaatje vanaf de eerste werkgroep, jou mag ik

natuurlijk niet missen. Bedankt voor inmiddels 11 jaar vriendschap. Wat goed dat we

elkaar niet uit het oog verliezen.

Ik heb nooit écht na hoeven denken over wie mijn paranimfen zouden worden.

Ik weet zeker dat ik met een stuk meer zelfvertrouwen op het podium zal staan met

Jennifer en Wessel aan mijn zijde. Lieve Jennifer, luister, we werken al twee jaar niet

meer samen en ik mis je nog steeds een beetje op de werkvloer. De dagen zijn eigenlijk

gewoon een stuk grappiger en leuker als jij er links- of rechtsom in voorkomt. Ik ben

dankbaar dat ik zo’n goede vriend heb ontmoet, daar kom je je er in je leven maar

een handjevol van tegen. (En terwijl ik dit typ krijg ik een appje van je dat ik nu toch

wel echt een drukker moet gaan regelen...jajaja.) Jen, ik vind jou dus echt een te gek

mens. “Sommige dingen gaan vanzelf en ik ga er vanuit dat we maatjes blijven zoals

we nu zijn.” Lieve Wessel, Arne Jon vertelde me in Den Haag al dat er een gast in het

VUmc werkte die ik echt een keer moest ontmoeten. En wat had hij een gelijk. Ik wil je

voor zó veel bedanken, bijvoorbeeld voor de gezellige borrels en festivals, voor je wijze

lessen, voor jouw perceptie (en daar draait het leven toch om), voor het tonen van

de binnenbocht, voor het mij in vertrouwen nemen als je zelf iemand nodig hebt om

mee te sparren, voor de rondjes in het Amsterdamse bos, voor de zwem-lunches, voor

mikslik Vondelpark en natuurlijk voor het letterlijke hoogtepunt op Tenerife. Bedankt

je vriendschap.

Family time!

Lieve Elly, Elibart, Robert, Sanne, Julia, Rosemarijn, Annelies, Wytse, Wende, Saskia,

Jasper, Maud en Thian, bedankt dat jullie mij me zo snel thuis en welkom hebben

laten voelen. Elibart, bedankt voor je oprechte interesse in het onderzoek. Niet

186

Dan

kwoo

rd

veel (niet-medische) mensen vragen door nadat ik in één zin verteld heb waar mijn

onderzoek over gaat. De Nederlandse samenvatting heb ik daarom een beetje met jou

in mijn achterhoofd geschreven.

Lieve opa’s en ome Jack, fijnere familie had ik niet kunnen wensen. Bedankt voor

alle logeerpartijen, verjaardagen, familieweekenden, Zonneland, en gezamenlijk

zomervakanties. Bedankt dat jullie mij mede gemaakt hebben tot wie ik nu ben. Jullie

worden ontzettend gemist.

Lieve René en Tessa, haha jullie vinden dit vreselijk dat weet ik zeker. En daarom

ga ik nu even extra gas geven. René, bedankt voor een leven lang gezelligheid en

lachen, bedankt voor het roddelen en het dobbelen, en natuurlijk bedankt dat we even

huisgenoten konden zijn. Dat was toch best een bijzondere tijd. Lieve Tessa, Ferry, en

blonde draakjes Lena, Jelle en Sofie, het is fijn om te merken dat jullie het belangrijk

vinden dat Marieke en ik de kinderen genoeg zien. Ik vond het heel tof om op Aruba

zo en dan gebeld te worden en op de camera de verbazing bij Jelle te zien omdat ome

Koen in februari in zijn zwembroek bij een zwembad zit.

Lieve papa en mama, zonder jullie was dit natuurlijk allemaal nooit mogelijk

geweest. Jullie hebben het voor mij, René en Tessa vanzelfsprekend laten voelen dat

we konden gaan studeren. Ik ging even zes jaar studeren, René plakte er als verrassing

nog even een extra master, en dus 2,5 jaar studie, aan vast. Dit konden we allemaal

vrij doen zonder dat we ons zorgen hoefden te maken over wat dan ook. En de centen

die we zelf verdienden mochten we altijd gewoon netjes bij de barman inleveren.

Ik realiseer me wel degelijk wat een voorecht dit voor ons was. Daarnaast heeft het

altijd heel fijn gevoeld dat jullie nooit druk op me gelegd hebben en dat jullie me altijd

onvoorwaardelijk gesteund hebben. Heel erg bedankt voor alles.

Lieve Marieke, wat ben ik blij dat René mij in maart 2015 meesleurde naar De Tuin

in Roosendaal. Bedankt voor je vertrouwen in ons, je geloof in mij, je geduld en je

steun. Bedankt voor alles wat je voor me hebt gedaan, bijvoorbeeld verhuizen naar

Amsterdam, mij lekker laten uitrazen als ik op vakantie plots tóch aan een rebuttal

moet beginnen, een frikadel speciaal onder mijn neus schuiven toen ik heimwee had

op Aruba, en natuurlijk het ontwerpen van de kaft én lay-out van dit proefschrift.

Bedankt voor al die mooie momenten die wel al beleefd hebben en voor alles wat nog

komen gaat. Lieve Marieke, met jou is alles leuker. Ik hou van je.

187

Ack

now

led

gem

ents

188

BIOGRAFIE

Biografie

Koen van Beers werd geboren op 3

juni 1988 te Roosendaal als zoon van

Jan en Marly van Beers, Samen met zijn

zus Tessa en broertje René groeide hij

op in Roosendaal. Hij ging daar naar de

basisschool (De Cortendijck), de middelbare

school (Het Gertrudiscollege) en werkte vele

jaren in Cafetaria De Schaopskooi. In 2006,

na het behalen van zijn Atheneum diploma,

startte hij zijn studie Geneeskunde aan het

Leids Universitair Medisch Centrum en

in 2008 verhuisde hij naar Den Haag. In

december 2012 behaalde hij zijn artsenbul

waarna hij in februari 2013 als ANIOS

Interne Geneeskunde in het Medisch

Centrum Haaglanden te Den Haag klinische

ervaring opdeed. In februari 2014 startte

hij zijn promotietraject bij de afdeling Interne Geneeskunde / Diabetes Centrum van

het VU medisch centrum, onder leiding van Professor Michaela Diamant, Professor

Frank Snoek, Dr. Erik Serné, en Dr. Nel Geelhoed-Duijvestijn waaruit dit proefschrift is

voortgekomen. Na het overlijden van Michaela Diamant werd Professor Mark Kramer

zijn eerste promotor. In maart 2015 verhuisde Koen naar Amsterdam en ontmoette hij

zijn vriendin Marieke. In januari 2017 vertrok hij, samen met Marieke, voor 8 maanden

naar Aruba om als zaalarts te gaan werken in het Dr. Horacio E. Oduber Hospitaal. In

september 2017 is Koen begonnen met de opleiding tot internist in het OLVG West te

Amsterdam (opleiders dr. M.C. Weijmer en prof.dr. Y.M. Smulders).

Bio

graf

ie

Sum

mar

y an

d D

iscu

ssio

n

189

Biography

Koen van Beers was born the 3th of June, 1988 in Roosendaal, the Netherlands as the

son of Jan and Marly van Beers. Together with his sister Tessa and brother René he grew

up in Roosendaal, went went to primary school (De Cortendijck) and high school (Het

Gertrudiscollege), and worked at cafeteria “De Schaopskooi” for many years. In 2006,

after graduating high school, he started his medical training at the Leiden University

Medical Center and in 2008 he moved to The Hague. In December 2012 he received

his Medical Degree and in February 2013 he started as a Senior House Officer at the

department of Internal Medicine at the Medical Center Haaglanden in The Hague.

In February 2014 he started his PhD project at the department of Internal Medicine /

Diabetes Center of the VU University Medical Center in Amsterdam under supervision

of Professor Michaela Diamant, Professor Frank Snoek, Dr. Erik Serné, and Dr. Nel

Geelhoed-Duijvestijn, which resulted in the present thesis. After Professor Michaela

Diamant passed away, Professor Mark Kramer became his promotor. In March 2015,

Koen moved to Amsterdam and met his partner Marieke. In January 2017 he moved to

Aruba, together with Marieke, to work as a Senior House Officer at the Dr. Horacio E.

Oduber Hospitaal. In September 2017 Koen started his residency in Internal Medicine

in OLVG West, Amsterdam (supervisors dr. M.C. Weijmer and prof. dr. Y.M. Smulders).

Bio

grap

hy

Continuous Glucose Monitoring · Continuous Glucose Monitoring IN CONTROL of Type 1 Diabetes...

Documents

Transcript of Continuous Glucose Monitoring · Continuous Glucose Monitoring IN CONTROL of Type 1 Diabetes...