Continuous glucose monitoring in children with type 1 diabetes effects on glycaemic control

(1)

Örebro University

School of Medicine

Degree project, 15 ECTS Örebro, Sweden

May 2016

Continuous glucose monitoring in

children with type 1 diabetes – effects

on glycaemic control

Version 2

Author

Andreas Udén, Student

Supervisor

(2)

Abstract

Introduction: Keeping a normal and even blood glucose level, measured as HbA1c, is one of the most important factors to avoid future complications of T1DM. During childhood

difficulties are seen in keeping an adequate metabolic control. Because of this many studies today are focusing on how to improve treatment for children with T1DM, among which continuous glucose monitoring, CGM, is a growing area. With CGM comes the possibility to more accurately measure blood-glucose throughout the day and consequently lead to an

improved treatment and lowered, HbA1c levels. Aims: The aim of this study was to investigate if the switch from self-monitoring of blood

glucose to CGM has an effect on diabetes treatment and its effect on HbA1c in children with T1DM.

Material and methods: All patients who were selected for CGM treatment between the years 2014-2016 were included in this study.

Medical variables were gathered from the participants’ medical records. HbA1c at the last visit before starting CGM was compared to HbA1c after revisit within 1,5-4,5 months. Participants were divided into 2 groups, one with baseline HbA1c > 60 mmol/mol and one with HbA1c < 60 mmol/mol. Results: In total 66 subjects were included in the study. The mean age was 12.5(4.6) years and the mean HbA1c before starting with CGM was 61.8(16.4) mmol/mol. The group with

baseline HbA1c > 60 mmol/mol had a significant decrease in HbA1c, 72.3(13.4) mmol/mol to 65.9(12.7) mmol/mol; p = 0.02. All participants combined had a decrease in HbA1c which was not significant; p = 0.06. Participants and parents reported that intervention of CGM had implemented an overall positive change to their quality of life. Parents felt less anxious about events of hypoglycaemia.

Conclusions: In conclusion, our results indicate that CGM could be a valid intervention for children with T1DM, both regarding metabolic control and psychosocial situation of the patients. Based on our study, no conclusions about the cost effectiveness of CGM could be drawn. Although our results indicated the benefits of CGM could lead to a decrease in expenses associated with both complications and social factors, further studies are needed to evaluate this aspect.

(3)

1. List of abbreviations

1. CGM = Continuous glucose monitor 2. HbA1c = Glycated haemoglobin 3. MDI = Multiple daily injections

4. SMBG = Self-monitoring of blood glucose 5. T1DM = Type 1 Diabetes Mellitus

6. SD = Standard Deviation

7. SAP = Sensor-augmented insulin pump therapy

8. CSII = Continuous Subcutaneous Insulin Infusion

9. CVD = Cardiovascular disease

10. MACE = Major Adverse Cardiovascular Events

(5)

2. Introduction

Type 1 diabetes (T1DM) is a disorder in which the insulin producing beta-cells of the pancreas are dysfunctional and eventually atrophied [1,2]. Insulin is a metabolic hormone which has a regulatory effect on blood glucose and essentially works to aid glucose in entering the cells and thereby lowering the blood glucose [19]. A loss of function of the beta-cells leads to a metabolic imbalance and ultimately an insufficient level of insulin in the blood. This results in an increased blood glucose level [19].

T1DM typically presents itself during early childhood until adolescence [3]. About 7,000 children-adolescents age 0-18 in Sweden have T1DM [21]. Adolescence is a period where the body goes through many critical stages of development and because of this is more sensitive to changes in the homeostasis [4,20]. An organ extra sensitive to homeostatic balance is the brain, which despite its relatively small size requires a large amount of the body energy. This applies especially to children, where at the age of 10, the brain peaks at an energy usage of about twice as high as an adult [5].

Known long term complications to T1DM as a consequence of prolonged hyperglycaemia are retinopathy leading to vision loss, peripheral neuropathy, atherosclerosis and kidney failure [6,7,21]. During puberty an increased risk for vascular damage is seen compared to young adults with T1DM [6]. Hyperglycaemia can also lead to a state called ketoacidosis which means the body pH is < 7,30 [7,21]. In the short term episodes of severe hypoglycaemia can lead to altered consciousness and ultimately unconsciousness with risk of damage to the vital organs and, if left untreated, death. Hypoglycaemia is also correlated with decreased cognitive ability [8, 9,10,11].

In conclusion this means an early, adequate treatment aimed to keep a stable blood glucose level is central to the development and outcome of the disease.

Treatment

The treatment of T1DM is based around two principles; measuring blood glucose level and injecting exogenous insulin to the blood. There are different methods on both monitoring blood glucose as well as the intake of exogenous insulin and during the last decade many new methods for treatment has developed [20].

Traditionally the method on exogenous insulin-intake has been through multiple subcutaneous injections, (MDI) where the insulin is absorbed into the blood stream to lower the glucose

(6)

level [12]. Although this is still considered a viable and still frequently used method to treat T1DM, more focus is switching to using insulin pumps [12]. Insulin pumps are devices that can either be attached into the subcutaneous layer of the lower abdomen or upper arm. They are loaded with multiple doses of insulin and are set to inject insulin at certain times during the day [12,20]. Insulin pumps can aid in treatment in multiple ways, for example a decreased risk of the patient forgetting to take his/her injection during certain times of the day as the pump is pre-set to give basal insulin during the day. The patient only has to manually select the pump to give bolus insulin before each meal [9,13].

Blood glucose levels can be measured by using different methods and by analysing several different variables, among which HbA1c has shown to be one of the most important factors when it comes to measuring the blood glucose over a long period of time. [7,14,20] HbA1c is the amount of glycated Haemoglobin, Hb, in the blood. As Hb is mostly bound to

erythrocytes with a life span of approximately 120 days, this is very accurate way of

measuring the average blood glucose over time [14,20]. Blood glucose can also be monitored by measuring the current blood glucose level, p-glucose at a certain point of time, called self-monitoring of blood glucose (SMBG) [7,14,20]. While both of these are viable factors to measure for T1DM treatment, p-glucose test only takes into account the glucose level at one point in time whereas HbA1c looks at the average glucose value over a few months, giving a more accurate answer on how well the diabetes is treated [14,20].

Different methods on how to monitor blood glucose have developed as well. Although the traditional way of monitoring by multiple times a day measuring blood glucose (SMBG) is still widely being used, another growing method is now available as an alternative or

complement [14,20]. This method is based around continuous glucose measurement with the aid of a device called CGM (continuous glucose monitor) [14,20]. A CGM is a device fixed to the arm, abdomen or outer part of bum with a small subcutaneous sensor, continuously

measuring the glucose level from the interstitial fluid which can either be scanned with a portable device or transferred wirelessly to a monitor giving the user the opportunity to see a dynamic view of current blood glucose level and how it has changed over the day [14,20,23]. There are different types of CGMs, a few exampes are Freestyle Libre, Dexcom and

Guardian. These monitors differ somewhat in how they function; Dexcom has no age limit, has an alarm system which alerts the user if blood glucose is too high/low. Dexcom needs to be calibrated through SMBG every 12 hours. Data from Dexcom is continuously transferred over to a portable monitor[24]. Freestyle Libre on the other hand has a 4-year age limit, no

(7)

alarm system but does not require regular calibration. Data from Libre is scanned with a portable monitor [25].

Measuring blood glucose continuously through CGM can be used as a tool to more accurately measure blood glucose and therefore possibly more adequately treat diabetes and by that also lower HbA1c. A correlation between use of CGM and improved treatment with lowered HbA1c has been shown in several earlier studies [13,15,16,17]. Other studies have shown that patients with CGM feel better about their treatment compared to those using SMBG

[16,18,17].

However, the potential of a more accurate treatment comes at a cost: CGM is more expensive and costs approximately 14,000 SEK (Freestyle libre) to 28,000 SEK (Dexcom) more per year than monitoring with only SMBG [20]. Because of this, much effort is today being put into finding out what effects CGM has on diabetes treatment and if it could be a viable method worth investing in despite being expensive. Many studies are trying to find if an improved treatment early on would potentially decrease the amount of complications later in life [16,17,18].

The aim of this study was therefore to investigate if the switch from SMBG to CGM has an effect on diabetes treatment and its effect on HbA1c in children with T1DM.

3. Materials and methods

3.1 Study design

The study is a retrospective review of medical records based on material at Pediatric department at university hospital Örebro.

3.2 Participants

All patients, a total of 66, who were selected for CGM treatment between the years 2014-2016 were included in this study.

The participants were selected for CGM based on specific indications, these indications were: 1. More than 10 measurements of P-glucose per day (SMBG)

(8)

2. HbA1c over 70 mmol/mol 3. Below 7 years of age 4. Puberty Tanner stage 2-4

5. Cognitive and/or neuropsycatric disorder 6. Family 2 or more children with T1DM

7. 2 or more severe cases of hypoglycaemia a year

In total 200 children and adolescents with T1DM are treated at the Department of Pediatrics Örebro. Their mean age is 12.9 (4.0) years and 58% are treated with CSII. Mean HbA1c is 57.6 mmol/mol and only 6% of the population has a HbA1c over 70 mmol/mol.

3.3 Method

Certain variables were selected to be assessed from the participants’ medical record. The time period reviewed was between 2014-2016. A comparison betweenbaseline “before-HbA1c”-values and “after-HbA1c” after revisit within 1,5-4,5 months was made. Some participants had more than 1 revisit during this period and consequently had their “after-HbA1c” based on the mean HbA1c of those visits.

This study included patients with Freestyle Libre and Dexcom.

Information regarding quality of life was gathered from medical records as well. This was done by reviewing the medical records of the revisits after start of CGM. Patients and parents were asked what they thought about the CGM if it worked well for them.

Variables:

Most variables gathered were used to receive the baseline values for the patient groups. These variables were age, gender, duration of diabetes, HbA1c, Standard Deviation of p-glucose measured in mmol/mol (SD), BMI. If subjects were using an insulin pump and if they were counting carbohydrates were also included in the baseline values, both given as a percentage of the whole group.

(9)

3.4 Statistics

SPSS was used as a program for statistics. A paired T-test was used to specifically analyse the change in the value of HbA1c among other variables over time. A chi-square test will also be used to analyse category variables.

To compare “before HbA1c” and “after HbA1c” the participants were divided into a group with HbA1c above 60 mmol/mol and one with HbA1c below 60 mmol/mol. The whole group was divided into these two groups because the median HbA1c value of all participants was 60 mmol/mol.

3.5 Ethics

The data has been gathered through the subject’s medical records. In order to access the records, permission was needed from the director of the paediatric department. The patients’ social security number has also been decoded and given a number before they were put in the document. The anonymous data were locked safely in a file. The question formulations were limited so no unnecessary data would be assessed in the clinical records of the participants.

4. Results

4.1

Baseline characteristics

A total of 66 participants were included is this baseline test. No subjects stopped with CGM entirely, but a few switched over to a different type. Mean age of the group was 12.5 (4.6) years. The average HbA1c of the group was 61.8 (16.4) mmol/mol. Of all participants, 58% were using an insulin pump.

A significant difference in mean age between girls and boys was seen; mean age of girls was 13.8 (3.5) yearsand boys 11.1 (5.2) years; p < 0.05. A tendency to a difference in BMI was also observed between the sexes.

Table 1. Clinical characteristics at baseline

Girl Boy Total

Number 35 31 66

Age (years)

n=66

(10)

Duration of diabetes (years) n=66 5.6 (4.1) 3.9 (3.0) 4.8 (3.9) P=0.08 HbA1c (mmol/mol) 60.7 (16.2) n=32 63.1(16.8) n=30 61.8 (16.4) n=62 ns SD (mmol/l ) 4.9 (1.1) n=18 4.4 (1.6) n=16 4.7 (1.4) n=34 ns BMI (kg/m2₎ _23.5_(6.4) n = 28 20.4 (5.8) n = 28 22.0 (6.2) n = 56 P=0.06 Pump (%) 54 % 61 % 58 % ns Carbohydrate counting (%) 46 % 64 % 54 % ns

Data are presented as mean ± SD for normally distributed variables and as percent for categorical variables. The amount of participants in each variable is given by “n”. Un-paired t-test or Chi square test were used for comparisons between genders.

4.2

Indications for CGM

The table below shows that the most common indication for CGM was puberty. Many p-glucose controls and high HbA1c were also important indications. More girls were selected based on puberty. It was more common for boys to be selected because of high HbA1c. Table 2. Indications for CGM/FGM

Girl Boy Total

Over 10 p-glucose per day (SMBG) 10 10 20

HbA1c more than 70 mmol/mol 7 13 20

Under 7 years of age 3 7 10

Puberty Tanner stage 2 - 4 25 17 42

Cognitive disorder 1 2 3

Family w/ or more children w/ T1DM 3 0 3

2 or more severe hypoglycaemias per year 0 0 0

Some subjects had multiple indications, hence the total amount of selected indications is higher than the amount of participants.

(11)

4.3

Effects on HbA1c

As is shown below, the group with baseline HbA1c above 60 mmol/mol had a significant decrease 72.3 mmol/mol to 65.9 mmol/mol; p = 0.02.

The group with HbA1c below 60 mmol/mol as well as the two groups combined did not have a significant decrease in HbA1c.

Table 3. Effect of CGM/FGM after 1,5- 4,5 months

HbA1c < 60 mmol/mol HbA1c > 60 mmol/mol Total

Before After Before After Before After

HbA1c (mean ± SD) 50.7 (5.8) n=14 52.2 (6.4)A n=14 72.3 (13.4) n=22 65.9 (12.7)B n=22 63.9 (15.4) n=36 60.6 (12.7)C n=36

Data are presented as mean ± SD. The groups were compared with a paired t-test. A: p = 0,31; B: p = 0,02;C: p = 0,06

There was no significant difference observed between boys and girls regarding the change of HbA1c-value.

Also, there was no correlation seen between age and decrease in HbA1c. When compared subjects treated with insulin pumps to those who did not have one, no significant difference was seen between the two regarding HbA1c decrease.

(12)

4.4 Correlation Baseline HbA1c and Delta HbA1c

Fig.1. A significant correlation was found between baseline HbA1c and its change on revisit. With higher baseline values a bigger decrease in HbA1c on revisit was seen. Pearson

coefficient r = 0.58; p < 0.01.

4.5 Quality of life

Most subjects reported to be pleased with their CGMs. The biggest change it implemented on both patients and family was reducing the fear of hypoglycaemia. Some parents reported to feel less anxious about hypoglycaemia during the night when their child had CGM.

(13)

5. Discussion

In this study the aim was to see what effects the intervention of continuous glucose

monitoring had on T1DM treatment and specifically the effects it had on HbA1c. What we found was that the group with high HbA1c, over 60 mmol/mol had a significant decrease in HbA1c 72.3 (13.4) mmol/mol to 65.9 (12.7) mmol/mol.

Earlier studies have investigated what effects CGM has on T1DM looking at both treatment and on everyday life dealing with the disease [16,17,18]. The results have been mixed showing a general decrease of HbA1c-value in the adult age-group, while it among the children/adolescents group has been harder finding significant data showing a clear

correlation between CGM and lowering of HbA1c [18]. It was observed however, that this difference had to do with the use of CGM, where the adult-group used their CGM much more frequently than the children/adolescents did [16,18].

When instead dividing the subjects into a group who used CGM at least 6 times/w and one who did not data showed a significant decrease in HbA1c in the group who used it more frequently. This indicates that when used regularly, CGM could prove to be effective in lowering HbA1c even in the younger population [16,22].

These studies also investigated how continuous glucose monitors affected patients and relatives in everyday life. What they found was almost exclusively positive reactions from both the patients and families [16,18].

This could be seen in our study as well, where the majority of the patients were pleased with their CGM. Both patients and relatives felt less worried regarding hypoglycaemia and in general. This observation could be of importance because some patients did not treat their diabetes adequately due to the fear of “over-treating” with insulin resulting in hypoglycaemia. What this means is that introducing CGMs to these patients could potentially help with

overcoming this threshold and thus resulting in a better treatment. Another mechanism to explain the improved HbA1c could be that the subjects, thanks to the continuous

measurements of glucose with the CGM, gathered more information about daily glucose values and eventual fluctuations. This would give more opportunities to modify treatment and more quickly compensate for high/low glucose-values during the day. In this way a lower and more stable HbA1c could be achieved.

(14)

An important observation based on these results is that the group with HbA1c > 60mmol/mol had a mean-value considered to be in the risk zone > 70mmol/mol [21]. If HbA1c is left at >70mmol/mol for a longer period of time the risk for long term complications increases, which is why it is of great importance to lower the HbA1c of these patients. After

intervention with CGM the mean HbA1c of this group decreased below 70 mmol/mol, greatly decreasing the risk for future complications. Considering the fact that this study was based on children/adolescents, which is an age-group very sensitive to metabolic imbalances, the urge to reach lower HbA1c levels is even higher.

We also investigated to see if there was a correlation between the baseline HbA1c-value and size of change in HbA1c (Delta HbA1c) on revisit. What we found was that the higher the baseline value HbA1c was, the larger was Delta HbA1c indicating that at high HbA1c values CGM has most of its effects, further supporting our results that CGM works best on patients with HbA1c > 60 mmol/mol

Participants with HbA1c below 60 mmol/mol did not show a decrease, in fact, a slight increase was seen. However, this increase was not significant; p = 0,31. The low amount of participants could be one of the reasons to this. Another possible cause is that the baseline-value of the group was within the recommended guidelines for metabolic control meaning a slight deviation of the HbA1c could be expected [21].

Regarding the group with HbA1c below 60 mmol/mol, the reasoning behind intervention with CGM for them would rather be focused on quality of life-factors than improving HbA1c, as this group already was within the recommended HbA1c values. The whole situation around the child with T1DM can be very tough. For example, fatigue syndrome among parents to children with T1DM is not an uncommon effect from months or even years of intense

treatment [26]. Another factor which could be contributing to fatigue is the near-constant fear of some parents that their child could turn hypoglycaemic, especially during night [23,27,28]. This consequently leads to many parents “taking shifts” being awake and watching over their child during the whole night, every day[23]. Some parents in our study even had to long term sick leave from work in order to recover from their fatigue syndrome condition. However, with the technology of CGM, the inbuilt alarm function gives users the opportunity to select certain glucose values as “threshold” when glucose levels are too high or too low. In our study this function was seen as a major benefit especially among the parents as many of the affected reported to feel a pronounced improvement of recover and sleep.

(15)

Regarding the cost of CGM compared to SMBG, multiple factors need to be accounted for. Although CGM today is considerably more expensive than measuring with SMBG only, the potential improvement in treatment and thereby HbA1c could have a pronounced effect on long term complications of diabetes.

A study investigated to see if HbA1c values in mid-age T1DM patients could be correlated to the risk for MACE and death from cardiovascular events after planned CABG (Coronary Artery Bypass Grasp) operations[29]. Patients were divided into groups based on HbA1c levels, ranging from > 53 to < 86 mmol/mol. They found a significant correlation; p < 0.05 showing an increased risk of about 18% for every 11 mmol/mol increase in HbA1c,

independent of earlier HbA1c level. This suggests the risk of late complications due to poorly

treated increase along with HbA1c [29]. From this it stands clear that an improvement in HbA1c conversely would lead to less long

term complications and therefore a pronounced improvement for the patients’ well-being and health. This in turn would also mean less resources would be required by the healthcare, and thereby reduced expenses.

As discussed earlier, the aspect of quality of life around the child and parents has a high value as well. For example, if CGM could decrease the stress around the child with his/her disease this would potentially result in less parents suffering from fatigue syndrome and because of this an overall improvement would be seen concerning the parents’ health. As a consequence, there would possibly be a decrease in expenses for society because less parents would be long term debilitated from their workplace. However, there has not been much research done on cost effectivity of CGM, so further studies would be needed to support this potential correlation, as of now, it is more a hypothesis.

A limiting factor to this study could be the size of the material which could have had an effect on the outcome of this study. Almost half of the participants did not have a revisit within the interval 1,5-4,5 months, meaning 36 out of 66 remained for the comparison between the HbA1c before and after CGM-use.

More participants would increase the probability of potentially showing a correlation between CGM-use and improved HbA1c-values. It would had been interesting in future studies to see what effect CGM had on girls and boys separately, and if there would be any differences among the two groups.

(16)

A difference between at what age the of boys and girls initiated their CGM treatment, where the boys did so at a mean age of 11.1 (5.2) and the girls at 13.8 (3.5) years old. It would be of interest to do further research on this as well to investigate why this difference exist.

Some of the chosen variables had too little data to be used in the comparison between baseline values and values at revisit. Among these were SD, which would be interesting to analyse in future studies to see if CGM had an effect on the fluctuation between high/low values of glucose. Results from earlier studies are suggesting there is a significant correlation between a decrease in SD with CGM in combination with SAP [15]. Furthermore, the study showed that removal of CGM was correlated with a decline in metabolic control.

In conclusion, our results indicate that CGM could be a valid intervention for children with T1DM, both regarding metabolic control and psychosocial situation of the patients. Based only on our study, no conclusions about the cost effectiveness of CGM could be drawn. Although our results indicated the benefits of CGM would lead to a decrease in expenses associated with both complications and social factors, further studies would be needed to evaluate this aspect.

(17)

6. Acknowledgements

Firstly, I would like to express my gratitude towards my supervisor Stefan Särnblad. Thank you for your constant support and encouragement throughout this period and for always being available whenever I wanted to discuss or ask something. I would also like to thank Lars-Göran Holmgren, nurse at Department of Pediatrics for your help and support during the collection of data and for introducing me to the concept of CGM.

(18)

7. References

1. Khodaeian M, Enayati S, Tabatabaei-Malazy O, Amoli MM. Association between Genetic Variants and Diabetes Mellitus in Iranian Populations: A Systematic Review of Observational Studies. J Diabetes Res 2015;2015:585917.

2. Turpaev K, Welsh N. Aromatic malononitriles stimulate the resistance of insulin-producing beta- cells to oxidants and inflammatory cytokines. Eur J Pharmacol 2016 May 10.

3. Fortunato F, Cappelli MG, Vece MM, Caputi G, Delvecchio M, Prato R, et al. Incidence of Type 1 Diabetes among Children and Adolescents in Italy between 2009 and 2013: The Role of a Regional Childhood Diabetes Registry. J Diabetes Res 2016;2016:7239692.

4. Hofer SE, Raile K, Frohlich-Reiterer E, Kapellen T, Dost A, Rosenbauer J, et al. Tracking of metabolic control from childhood to young adulthood in type 1 diabetes. J Pediatr 2014 Nov;165(5):956-61.e1-2.

5. Arbelaez AM, Semenkovich K, Hershey T. Glycemic extremes in youth with T1DM: the structural and functional integrity of the developing brain. Pediatr Diabetes 2013 Dec;14(8):541-553.

6. Cho YH, Craig ME, Donaghue KC. Puberty as an accelerator for diabetes complications. Pediatr Diabetes 2014 Feb;15(1):18-26.

7. Nathan DM, DCCT/EDIC Research Group. The diabetes control and complications trial/epidemiology of diabetes interventions and complications study at 30 years: overview. Diabetes Care 2014;37(1):9-16.

8. Ryan CM, Atchison J, Puczynski S, Puczynski M, Arslanian S, Becker D. Mild hypoglycemia associated with deterioration of mental efficiency in children with insulin-dependent diabetes mellitus. J Pediatr 1990 Jul;117(1 Pt 1):32-38.

9. Asvold BO, Sand T, Hestad K, Bjorgaas MR. Cognitive function in type 1 diabetic adults with early exposure to severe hypoglycemia: a 16-year follow-up study. Diabetes Care 2010 Sep;33(9):1945-1947.

10. Gaudieri PA, Chen R, Greer TF, Holmes CS. Cognitive function in children with type 1 diabetes: a meta-analysis. Diabetes Care 2008 Sep;31(9):1892-1897.

(19)

11. Hannonen R, Komulainen J, Riikonen R, Ahonen T, Eklund K, Tolvanen A, et al. Academic skills in children with early-onset type 1 diabetes: the effects of diabetes-related risk factors. Dev Med Child Neurol 2012 May;54(5):457-463.

12. Tumminia A, Crimi S, Sciacca L, Buscema M, Frittitta L, Squatrito S, et al. Efficacy of real-time continuous glucose monitoring on glycaemic control and glucose

variability in type 1 diabetic patients treated with either insulin pumps or multiple insulin injection therapy: a randomized controlled crossover trial. Diabetes Metab Res Rev 2015 Jan;31(1):61-68.

13. Slover RH, Welsh JB, Criego A, Weinzimer SA, Willi SM, Wood MA, et al. Effectiveness of sensor-augmented pump therapy in children and adolescents with type 1 diabetes in the STAR 3 study. Pediatr Diabetes 2012 Feb;13(1):6-11.

14. Rewers MJ, Pillay K, de Beaufort C, Craig ME, Hanas R, Acerini CL, et al. ISPAD Clinical Practice Consensus Guidelines 2014. Assessment and monitoring of glycemic control in children and adolescents with diabetes. Pediatr Diabetes 2014 Sep;15 Suppl 20:102-114.

15. Battelino T, Conget I, Olsen B, Schutz-Fuhrmann I, Hommel E, Hoogma R, 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

Dec;55(12):3155-3162.

16. Mauras N, Beck R, Xing D, Ruedy K, Buckingham B, Tansey M, 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 Feb;35(2):204-210.

17. Ludvigsson J, Hanas R. Continuous subcutaneous glucose monitoring improved metabolic control in pediatric patients with type 1 diabetes: a controlled crossover study. Pediatrics 2003 May;111(5 Pt 1):933-938.

18. Juvenile Diabetes Research Foundation Continuous Glucose Monitoring Study Group, Beck RW, Lawrence JM, Laffel L, Wysocki T, Xing D, 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 Oct;33(10):2175-2177.

19. Hall JE, Guyton AC. Guyton and Hall textbook of medical physiology. 12th ed. ed. Philadelphia, PA: Saunders/Elsevier; 2011.

(20)

20. SBU. Kontinuerlig subkutan glukosmätning vid diabetes Stockholm: Statens beredning för medicinsk utvärdering (SBU); 2013. SBU Alert-rapport nr 2013-04. ISSN 1652-7151.

21. SWEDIABKIDS. Årsrapport 2014. Nationellt register för barn- och ungdomsdiabetes. ISSN 2001-3701.

22. Juvenile Diabetes Research Foundation Continuous Glucose Monitoring Study Group, Beck RW, Buckingham B, Miller K, Wolpert H, Xing D, et al. Factors predictive of use and of benefit from continuous glucose monitoring in type 1 diabetes. Diabetes Care 2009 Nov;32(11):1947-1953.

23. Riemsma R, Corro Ramos I, Birnie R, Buyukkaramikli N, Armstrong N, Ryder S, et al. Integrated sensor-augmented pump therapy systems [the MiniMed(R) Paradigm Veo system and the Vibe and G4(R) PLATINUM CGM (continuous glucose monitoring) system] for managing blood glucose levels in type 1 diabetes: a systematic review and economic evaluation. Health Technol Assess 2016 Feb;20(17):1-252.

24. Christiansen M, Bailey T, Watkins E, Liljenquist D, Price D, Nakamura K, et al. A new-generation continuous glucose monitoring system: improved accuracy and reliability compared with a previous-generation system. Diabetes Technol Ther 2013 Oct;15(10):881-888.

25. Bailey T, Bode BW, Christiansen MP, Klaff LJ, Alva S. The Performance and

Usability of a Factory-Calibrated Flash Glucose Monitoring System. Diabetes Technol Ther 2015 Nov;17(11):787-794.

26. Lindstrom C, Aman J, Norberg AL. Increased prevalence of burnout symptoms in parents of chronically ill children. Acta Paediatr 2010 Mar;99(3):427-432.

27. Sarnblad S, Berg L, Detlofsson I, Jonsson A, Forsander G. Diabetes management in Swedish schools: a national survey of attitudes of parents, children, and diabetes teams. Pediatr Diabetes 2014 Dec;15(8):550-556.

28. Streisand R, Swift E, Wickmark T, Chen R, Holmes CS. Pediatric parenting stress among parents of children with type 1 diabetes: the role of self-efficacy,

responsibility, and fear. J Pediatr Psychol 2005 Sep;30(6):513-521.

29. Nystrom T, Holzmann MJ, Eliasson B, Kuhl J, Sartipy U. Glycemic Control in Type 1 Diabetes and Long-Term Risk of Cardiovascular Events or Death After Coronary Artery Bypass Grafting. J Am Coll Cardiol 2015 Aug 4;66(5):535-543.