© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.
Pediatric Diabetes 2015: 16: 256–262 doi: 10.1111/pedi.12164 All rights reserved
Pediatric Diabetes
Original Article
Insulin pump treatment; increasing prevalence, and predictors for better metabolic outcome in Danish children and adolescents with type 1 diabetes Olsen B., Johannesen J., Fredheim S., Svensson J. Insulin pump treatment, increasing prevalence, and predictors for better metabolic outcome in Danish children and adolescents with type 1 diabetes. Pediatric Diabetes 2015: 16: 256–262. Aims: Few studies have looked at nationwide data for insulin pump treatment. Since 1996 the Danish Childhood Diabetes Registry (DanDiabKids) has collected data on all Danish diabetic patients aged 0–15 yr. The purpose of this study is to evaluate the prevalence of continuous subcutaneous insulin infusion (CSII) use among Danish children with diabetes and to compare metabolic control in CSII-treated children and adolescents to those treated with MDI. Materials and methods: The Registry collects on a yearly basis data on insulin regimen, central measured hemoglobin A1c (HbA1c), and demographic data on all patients. In the period 2005–2011, 2983 young patients (1721 males) with diabetes were followed in the Registry. Mean observation period was 5.11 yr [standard error (SE) 0.09]. In the total period 1846 patients were treated with MDI and 1493 changed from MDI to CSII. In 2005, less than 5% of children were treated with CSII whereas the percentage of children on CSII increased to approximately 50% in 2011. The patients were divided into age groups, 15 yr. Results: HbA1c was significantly higher in MDI-treated children, +5.29 (CI 95% 4.29; 6.29 mmol/mol). HbA1c in all age groups was significantly lower in CSII-treated patients, and longitudinally HbA1c continued to be lower in all age groups. In multivariate analysis, a low HbA1c at CSII start, centers with more than 100 pump patients, a more recent year of diabetes onset, a higher number of self-monitoring of blood glucose (SMBG) measurements, a higher number of daily boluses, and a higher percentage of bolus insulin were all related to a lower HbA1c. Conclusion: The percentage of children on pumps (CSII) is CSII treatment is associated with a significantly lower Hba1c, achieved just after treatment initiation. In the following years there is a parallel rise in HbA1c in both MDI as well as in MDI treated patients. Patients coming from larger clinics, and patients measuring more blood glucose values and taking more boluses have a better metabolic control.
The Diabetes Control and Complications Trial Study demonstrated that intensive insulin therapy improves metabolic control and reduces the risk of microvascular complications in adults as well as adolescents with type 1 diabetes (T1D) (1, 2). It is well-known that a poor
256
B Olsena , J Johannesena,b , S Fredheima,b , J Svenssona,b and The Danish Society for Childhood and Adolescent Diabetes a
Department of Paediatrics, Herlev University Hospital, Herlev, Denmark; and b Faculty of Health and Medical Sciences, Copenhagen University, Copenhagen, Denmark
Key words: children and adolescents – HbA1c – insulin pump treatment – registry – T1D Corresponding author: Birthe Olsen Department of Paediatrics, Herlev University Hospital, Nordre Ringvej, 2730 Herlev, Denmark. Tel: +45 23204896; fax: +45 38685101; e-mail: [email protected] Submitted 20 December 2013. Accepted for publication 19 May 2014
metabolic control during the pre-pubertal period also contributes to the risk of complications, however to a lesser extent compared to the years after the initiation of puberty (3). However, due to several factors, such as high insulin sensitivity, hypoglycaemia unawareness
Insulin pump treatment in young Danish patients with diabetes and unpredictable eating and activity it is often a challenge to obtain a tight blood glucose control in very young children with T1D. During the last decade, the use of continuous subcutaneous insulin infusion (CSII) has become increasingly prevalent among children and adolescents with T1D. CSII is the most physiologic method of insulin delivery currently available, and is able to cope with changing insulin needs throughout childhood and adolescence (4). To prevent long-term micro- and macrovascular complications to diabetes, it is essential to achieve an optimal metabolic control already in childhood and adolescence. It is also important to avoid recurrent episodes of hypoglycemia, which, especially in young children seem to permanently affect certain areas of the cognitive function in the brain (5). There is, however, still controversy on the subject, and not all studies report cognitive damage due to hypoglycemia (6). However, not all children benefit from CSII and to improve outcome it is therefore important to identify individual factors that predict an improved glycemic control in young patients on CSII treatment. A recent study from DanDiabKids showed that metabolic control in Danish diabetic children improved significantly in the period 1997–2006, but with huge variation in metabolic control within Danish centers (7). The Danish study also showed that the national reduction in hemoglobin A1c (HbA1c) was associated to an increased number of self-monitored blood glucose (SMBG) measurements and a decrease in the number of hypoglycemic events in those with the best metabolic control. However, the majority of young Danish patients with diabetes do not reach the treatment target of HbA1c level below 7.5% (58 mmol/mol) (7). The purpose of this study is to evaluate the prevalence of CSII use among Danish children with diabetes and to compare metabolic control in CSII-treated children and adolescents to those treated with Multiple Daily Injection Therapy (MDI). Further to investigate predictors for better metabolic control in a nationwide population of CSII-treated patients.
Materials and methods This is a national registry-based study. Initially all children treated with insulin pumps were allocated to the case group; secondly a control group was selected from the remaining children in the population who stayed on pen treatment. The controls were included if there was a child in the case group with the same age (5-yr age groups), duration (1-yr intervals), and year of onset (3-yr intervals). In Denmark, all children with diabetes are taken care of by multidisciplinary teams, consisting of pediatric Pediatric Diabetes 2015: 16: 256–262
diabetologists, diabetes nurses, dieticians, and pediatric psychologists. The patients are seen in the outpatient clinics at least three to four times a year, and we have national guidelines for treatment and education of children with diabetes and their families. According to our guidelines all children are educated in carbohydrate counting (CHO) and are recommended to use it. However, at the moment we not have data in the registry on the actual use of CHO. The target for HbA1c for children and adolescents with diabetes is 3 months) have not been able to show a sustained significant benefit in terms of a lowering HbA1c in CSII-treated children. In order to select patients for CSII treatment it is thus very important to identify factors that predict or affect glycemic control during pump therapy. Since 2005, there has been an extreme increase in the number of Danish children with T1D treated with CSII, and this large register study, following patients for more than 5 yr, shows that children and adolescents in all age groups achieve improved metabolic control on CSII treatment. This is consistent with another study, following 421 children and young adults with T1D for 4 yr after CSII initiation (9). The study demonstrated a significant sustained decrease in HbA1c without increased rates of hypoglycemia, and achievement of target HbA1c was significantly associated with lower HbA1c, younger age (70 10% increase in fast-acting insulin Units of Insulin per kg HbA1c before pump treatment Number of boluses* bolus/d 2–3 boluses/d 4–5 boluses/d 6–7 boluses/d 8–10 boluses/d >10 boluses/d
HbA1c mmol/mol
p-value 0.20
1.52 (−1.01; 4.06) 1.56 (−0.05; 3.18) 1.54 (0.08; 3.00) 0 −0.07 (−1.01; 0.87) 0 −0.42 (−2.71; 1.86) −8.80 (−23.2; 5.57) −9.08 (−12.8; −5.40) −4.48 (−5.29; −3.66) 0
0.89 0.72
Pediatric Diabetes 2015: 16: 256–262 doi: 10.1111/pedi.12164 All rights reserved
Pediatric Diabetes
Original Article
Insulin pump treatment; increasing prevalence, and predictors for better metabolic outcome in Danish children and adolescents with type 1 diabetes Olsen B., Johannesen J., Fredheim S., Svensson J. Insulin pump treatment, increasing prevalence, and predictors for better metabolic outcome in Danish children and adolescents with type 1 diabetes. Pediatric Diabetes 2015: 16: 256–262. Aims: Few studies have looked at nationwide data for insulin pump treatment. Since 1996 the Danish Childhood Diabetes Registry (DanDiabKids) has collected data on all Danish diabetic patients aged 0–15 yr. The purpose of this study is to evaluate the prevalence of continuous subcutaneous insulin infusion (CSII) use among Danish children with diabetes and to compare metabolic control in CSII-treated children and adolescents to those treated with MDI. Materials and methods: The Registry collects on a yearly basis data on insulin regimen, central measured hemoglobin A1c (HbA1c), and demographic data on all patients. In the period 2005–2011, 2983 young patients (1721 males) with diabetes were followed in the Registry. Mean observation period was 5.11 yr [standard error (SE) 0.09]. In the total period 1846 patients were treated with MDI and 1493 changed from MDI to CSII. In 2005, less than 5% of children were treated with CSII whereas the percentage of children on CSII increased to approximately 50% in 2011. The patients were divided into age groups, 15 yr. Results: HbA1c was significantly higher in MDI-treated children, +5.29 (CI 95% 4.29; 6.29 mmol/mol). HbA1c in all age groups was significantly lower in CSII-treated patients, and longitudinally HbA1c continued to be lower in all age groups. In multivariate analysis, a low HbA1c at CSII start, centers with more than 100 pump patients, a more recent year of diabetes onset, a higher number of self-monitoring of blood glucose (SMBG) measurements, a higher number of daily boluses, and a higher percentage of bolus insulin were all related to a lower HbA1c. Conclusion: The percentage of children on pumps (CSII) is CSII treatment is associated with a significantly lower Hba1c, achieved just after treatment initiation. In the following years there is a parallel rise in HbA1c in both MDI as well as in MDI treated patients. Patients coming from larger clinics, and patients measuring more blood glucose values and taking more boluses have a better metabolic control.
The Diabetes Control and Complications Trial Study demonstrated that intensive insulin therapy improves metabolic control and reduces the risk of microvascular complications in adults as well as adolescents with type 1 diabetes (T1D) (1, 2). It is well-known that a poor
256
B Olsena , J Johannesena,b , S Fredheima,b , J Svenssona,b and The Danish Society for Childhood and Adolescent Diabetes a
Department of Paediatrics, Herlev University Hospital, Herlev, Denmark; and b Faculty of Health and Medical Sciences, Copenhagen University, Copenhagen, Denmark
Key words: children and adolescents – HbA1c – insulin pump treatment – registry – T1D Corresponding author: Birthe Olsen Department of Paediatrics, Herlev University Hospital, Nordre Ringvej, 2730 Herlev, Denmark. Tel: +45 23204896; fax: +45 38685101; e-mail: [email protected] Submitted 20 December 2013. Accepted for publication 19 May 2014
metabolic control during the pre-pubertal period also contributes to the risk of complications, however to a lesser extent compared to the years after the initiation of puberty (3). However, due to several factors, such as high insulin sensitivity, hypoglycaemia unawareness
Insulin pump treatment in young Danish patients with diabetes and unpredictable eating and activity it is often a challenge to obtain a tight blood glucose control in very young children with T1D. During the last decade, the use of continuous subcutaneous insulin infusion (CSII) has become increasingly prevalent among children and adolescents with T1D. CSII is the most physiologic method of insulin delivery currently available, and is able to cope with changing insulin needs throughout childhood and adolescence (4). To prevent long-term micro- and macrovascular complications to diabetes, it is essential to achieve an optimal metabolic control already in childhood and adolescence. It is also important to avoid recurrent episodes of hypoglycemia, which, especially in young children seem to permanently affect certain areas of the cognitive function in the brain (5). There is, however, still controversy on the subject, and not all studies report cognitive damage due to hypoglycemia (6). However, not all children benefit from CSII and to improve outcome it is therefore important to identify individual factors that predict an improved glycemic control in young patients on CSII treatment. A recent study from DanDiabKids showed that metabolic control in Danish diabetic children improved significantly in the period 1997–2006, but with huge variation in metabolic control within Danish centers (7). The Danish study also showed that the national reduction in hemoglobin A1c (HbA1c) was associated to an increased number of self-monitored blood glucose (SMBG) measurements and a decrease in the number of hypoglycemic events in those with the best metabolic control. However, the majority of young Danish patients with diabetes do not reach the treatment target of HbA1c level below 7.5% (58 mmol/mol) (7). The purpose of this study is to evaluate the prevalence of CSII use among Danish children with diabetes and to compare metabolic control in CSII-treated children and adolescents to those treated with Multiple Daily Injection Therapy (MDI). Further to investigate predictors for better metabolic control in a nationwide population of CSII-treated patients.
Materials and methods This is a national registry-based study. Initially all children treated with insulin pumps were allocated to the case group; secondly a control group was selected from the remaining children in the population who stayed on pen treatment. The controls were included if there was a child in the case group with the same age (5-yr age groups), duration (1-yr intervals), and year of onset (3-yr intervals). In Denmark, all children with diabetes are taken care of by multidisciplinary teams, consisting of pediatric Pediatric Diabetes 2015: 16: 256–262
diabetologists, diabetes nurses, dieticians, and pediatric psychologists. The patients are seen in the outpatient clinics at least three to four times a year, and we have national guidelines for treatment and education of children with diabetes and their families. According to our guidelines all children are educated in carbohydrate counting (CHO) and are recommended to use it. However, at the moment we not have data in the registry on the actual use of CHO. The target for HbA1c for children and adolescents with diabetes is 3 months) have not been able to show a sustained significant benefit in terms of a lowering HbA1c in CSII-treated children. In order to select patients for CSII treatment it is thus very important to identify factors that predict or affect glycemic control during pump therapy. Since 2005, there has been an extreme increase in the number of Danish children with T1D treated with CSII, and this large register study, following patients for more than 5 yr, shows that children and adolescents in all age groups achieve improved metabolic control on CSII treatment. This is consistent with another study, following 421 children and young adults with T1D for 4 yr after CSII initiation (9). The study demonstrated a significant sustained decrease in HbA1c without increased rates of hypoglycemia, and achievement of target HbA1c was significantly associated with lower HbA1c, younger age (70 10% increase in fast-acting insulin Units of Insulin per kg HbA1c before pump treatment Number of boluses* bolus/d 2–3 boluses/d 4–5 boluses/d 6–7 boluses/d 8–10 boluses/d >10 boluses/d
HbA1c mmol/mol
p-value 0.20
1.52 (−1.01; 4.06) 1.56 (−0.05; 3.18) 1.54 (0.08; 3.00) 0 −0.07 (−1.01; 0.87) 0 −0.42 (−2.71; 1.86) −8.80 (−23.2; 5.57) −9.08 (−12.8; −5.40) −4.48 (−5.29; −3.66) 0
0.89 0.72
