Most children and young people (CYP) with type 1 diabetes present with the classic symptoms of polyuria, polydipsia, lethargy and weight loss. However, healthcare professionals do not always consider diabetes in the context of childhood, and many parents may suspect diabetes but, not wanting to believe this could happen to their child, will delay seeking medical help (Lowes et al, 2005). This is problematic, particularly considering the subsequent outcome of diabetic ketoacidosis, and potential death, if diabetes is untreated.
There are particular difficulties in diagnosing young children, particularly those <2 years of age. Parental concerns may be that the child is fretful, restless or irritable but, on questioning, a story of increased or constant milk feeding and heavy, soaking nappies will be obtained. Older children may present with secondary nocturnal enuresis. Ketoacidosis will cause Kussmaul breathing, leading to a diagnosis of a chest infection, or abdominal pain which may be mistaken for an acute surgical emergency. Urine testing for glucose and ketones is easy in most children (if they have diabetes they will be polyuric). If positive and/or a random capillary blood glucose is greater than 11.1 mmol/L an urgent discussion with the local paediatrician should take place (Alberti et al, 2004).
However, there are a number of other forms of diabetes that are increasingly being diagnosed (Table 1). Diabetes arises as part of cystic fibrosis, for example, and can occur in any child on high-dose steroid replacement – the highest prevalence is likely to be among those undergoing chemotherapy. In addition, other primary forms of diabetes occur, which is important because they may present with an incidental finding of hyperglycaemia or glycosuria in an otherwise well child (Porter and Barrett, 2004). Table 1 highlights the differential diagnosis of glucose intolerance in childhood and the likely clinical presentation. Of these, an increasing incidence of type 2 diabetes, probably related to the increased prevalence of childhood overweight and obesity, has caused the greatest concern (Matyka, 2008).
There are now more CYP presenting with type 2 diabetes, especially in areas with high proportions of ethnic groups. The numbers are still relatively small and we do not seem to have the same high prevalence as seen in North America or Japan (Matyka, 2008). The clinical presentation of CYP with type 2 diabetes can be similar to that of a child with type 1 diabetes, and cases with ketosis have been described although this is often mild (Fagot-Campagna et al, 2000; see Box 1 for a case report to illustrate this) (Table 1).
The majority of children with type 2 diabetes in the UK are from south Asian backgrounds and have a strong family history of type 2 diabetes (Haines et al, 2007). Making a confident diagnosis of type 2 over type 1 diabetes can be a challenge even for an experienced paediatrician. The most pragmatic approach in case of any doubt is to give insulin therapy (Rosenbloom et al, 2009): this is likely to be necessary where blood glucose levels are in double figures or in the presence of significant ketonaemia even in a person with type 2 diabetes. Once the child is safely on treatment and stabilised, the precise diagnosis can then be reviewed regularly in the relative safety of the diabetes clinic.
Cases of delayed or missed diagnoses of diabetes are regularly observed (Ali et al, 2011). About 30% of children with new onset diabetes have had at least one related medical visit prior to being diagnosed (Bui et al, 2010). The National Patient Safety Agency is now requesting that all such cases are reported as clinical adverse events. At best, children can still be relatively well at presentation; at worst, they can die before they reach the hospital. Where there is any concern about the possibility of diabetes in a CYP it is essential to discuss the individual with a senior paediatrician as soon as possible.
Nutrition and weight management for overweight children with diabetes
Most children with type 1 diabetes do not have a significant weight problem. However, the inability to deliver insulin in a physiological manner means that most will be over-insulinised (Holl et al, 1998; 2003), except perhaps those on insulin pump therapy, and thus will be heavier than their counterparts without diabetes.
There have been a number of changes to the nutritional management of people with type 1 diabetes over the years. Currently the emphasis is on healthy eating, and the International Society for Pediatric and Adolescent Diabetes (ISPAD) has produced guidelines which recommend that the composition of a healthy diet is 50–55% carbohydrate, of which sucrose can be up to 10% of total daily energy intake, 30–35% should be fat, of which less than 10% should come from saturated or trans fatty acid sources, and 10–15% should be protein (Smart et al, 2009). Salt intake should be less than 6 g per day, much of which is already present in the highly processed food consumed in the developed world.
These dietary recommendations apply across the paediatric age range but a flexible approach will be necessary for younger children, particularly toddlers. Frequent small meals, or “grazing”, are typical of young children and it will be easier to adapt the insulin regimen to fit the child rather than ask the parents to make drastic changes to the diet of their young children who already provide a number of challenges.
Studies have examined the diet of different age groups of CYP (Smart et al, 2009). There are methodological problems with examining dietary patterns in a group of people who have been given a lot of information on how to eat. Food diaries, often used as a gold standard for assessment, are also an effective deterrent to “normal” eating: when being asked to fill in a food diary for 3 days one may well reward oneself with a burger and chips or a large bar of chocolate on day four. However, clinical experience does suggest that people with diabetes are no different to the general population in terms of what may be called dietary “indiscretions”. Ideally these would be infrequent, and although blood glucose levels can be controlled by carbohydrate counting and adjusting insulin dosage, the extra calories can lead to significant weight gain, which can be difficult to lose.
It is likely that most people with type 2 diabetes will be either overweight or obese, however it is beyond the scope of this article to describe in detail the approaches to weight management for childhood obesity. A number of studies have been performed in CYP, varying from lifestyle adjustments to weight loss medication such as orlistat and sibutramine, and have been reviewed within a Cochrane review (Oude Luttikhuis et al, 2009) and NICE (2006) guidance is available. The majority of these studies have included small sample sizes and have been of short duration with many showing little effect on weight change.
There is no doubt that weight loss medications can be effective in sub-groups of people. A study of sibutramine involving just under 500 adolescents showed that at 12 months the estimated mean change in BMI from baseline for sibutramine was –3.1 kg/m2 versus –0.3 kg/m2 for placebo, and this difference was statistically significant at all study visits, even at week 1 of treatment (Berkowitz et al, 2006). Sibutramine never had a license in the UK for those aged under 18 years but has lost its license completely due to concerns over cardiovascular safety so this avenue of management has now been closed.
There do not appear to have been any studies that have specifically assessed weight management strategies in CYP with type 2 diabetes, but there is no reason to suspect that generic weight management approaches would not be appropriate. Children and their families will need to be motivated and engaged for any weight-loss strategy to be effective. Data suggest that this can be very difficult, and studies also show that families struggle to recognise that their child may have a significant weight problem (Etelson et al, 2003; Eckstein et al, 2006). This may be because many parents will compare their child’s physique with other children in the playground; with 20% of children overweight or obese in the UK, some parents may have false reassurance from playground comparisons (Etelson et al, 2003; Eckstein et al, 2006).
Worryingly, it is has been shown that even adolescents with type 2 diabetes and their parents underestimate the severity of the weight problem (Skinner et al, 2008). In one study the mean BMI of 104 adolescents with diabetes was 36.4 kg/m2; 87% were classed as overweight by North American guidelines (Barlow and Dietz, 1998). Yet 40% of parents whose child had a BMI >95th percentile felt that their child’s weight was “about right” and 55% of adolescents with a BMI >95th percentile were unconcerned by their weight. There was consistency in weight estimation between adolescents and their parents. Those parents and adolescents who underestimated weight problems had poorer dietary and physical activity behaviours than those who were more aware of the weight issues (Skinner et al, 2008). Adequate explanation of the importance of weight management for people with type 2 diabetes is therefore paramount.
Management of children with type 1 diabetes
The management of the CYP with diabetes will depend on a number of factors, including age, weight, developmental stage, cognitive ability and family support.
As in most areas of paediatrics, doses of medications, in this case insulin, are initially based on body weight. Doses prescribed at diagnosis are informed “estimates” of likely need. Most centres are likely to start doses at the lower end of the estimated range to avoid provoking hypoglycaemia soon after diagnosis. Most CYP will be quite insulin resistant for the first few weeks (approximately 4–8 weeks) after diagnosis due to the preceding persistent hyperglycaemia causing glucose toxicity just prior to diagnosis, and doses may need to be increased quite rapidly in the early stages. It is suggested that total daily insulin requirements will vary from 0.7–1.0 units/kg/day for pre-pubertal children (Bangstad et al, 2009).
For young people going through puberty the insulin requirements will go as high as 1.5–2.0 units/kg/day during the rapid growth phase, usually mid to late puberty. The increase in insulin requirements during the pubertal growth spurt is felt to be due to increases in growth hormone released during puberty resulting in insulin resistance (Dunger, 1992).
In addition, the insulin deficiency of type 1 diabetes leads to abnormalities in the growth hormone/insulin-like growth factor-1 (IGF-1) feedback mechanism and pubertal children over-secrete growth hormone making puberty a time of marked insulin resistance for children with type 1 diabetes (Dunger, 1992).
Total daily doses of insulin are likely to be very high during puberty as young people also significantly increase their calorie intake to fuel normal growth (a 60 kg boy could need up to 120 units per day), causing anxiety for these individuals and their carers alike. Despite these large doses of insulin, even adolescents who are very engaged with their diabetes management can struggle to achieve tight glycaemic control.
Data from the DCCT (Diabetes Control and Complications Trial) are quite old now but they do highlight the difficulties for adolescents. A total of 195 adolescents aged between 13 and 17 years participated in the trial (DCCT Research Group, 1994). Although the beneficial effect of improved glycaemic control was still evident, adolescents found it more difficult to achieve lower HbA1c levels than their adult counterparts (8.06±0.13% [64±1.4 mmol/mol] versus 7.12±0.03% [54±0.03 mmol/mol]; P<0.001), yet they had a greater tendency towards severe hypoglycaemia (85.7 events per 100 person years versus 56.9 events in the adult cohort) (DCCT Research Group, 1994).
Young children with diabetes are likely to need shorter needles than adolescents and adults, however this will need to be assessed with each individual – some teenagers can be very slim and the author has used short (4 mm) needles in this group often at diagnosis where there has been significant weight loss. All children are advised to rotate injection sites both throughout the day and from day to day to avoid the development of lipohypertrophy, although it is the author’s experience that there are a small number of people who appear to be particularly prone to this complication even with good site rotation.
There are few studies that have examined the optimum diabetes regimen for children at different ages in a structured experimental fashion (Spinks et al, 2007). As a result the choices of regimen and insulin type are likely to be a negotiation between child, family and their healthcare team.
Many clinical teams are recommending intensive insulin regimens, either multiple daily injections (MDI) or insulin pump therapy, for all CYP with diabetes (Gosden et al, 2010). This will involve carbohydrate counting and dose adjustment of mealtime rapid-acting insulin. These are likely to be the best regimens for very young children (under 5 years of age) who may eat unpredictably both with respect to timing and quantity of food. Rapid-acting insulin can be given after food for these children so the dose can be tailored to the amount of food actually eaten rather than the amount of food offered.
Using MDI is a good way of managing food refusal in any age of child without having to worry about the risk of hypoglycaemia. This is also a very good regimen for adolescents who will have much less structure to their lives and want to go out and socialise with friends at varying times. However, these regimens do involve a high treatment burden and require a great deal of individual and family engagement for them to work well – adolescence can be a very difficult time for this kind of treatment adherence. Many outpatient consultations with adolescents will involve negotiations about how to fit diabetes around the life of someone who would prefer to believe that it was not there.
There will be some families who will struggle with these regimens. Carbohydrate counting and insulin dose adjustment based on ambient glucose, food and physical activity are likely to require a reasonable level of understanding and can be extremely challenging for families with special educational needs (either in the child or supporting family) or problems with literacy or numeracy. Support is available, often coming from the local diabetes team from the specialist nurse, but some families are still left with little confidence. In the end, people with diabetes have to make decisions about their diabetes management almost continuously and no diabetes care team will be able to provide this level of input.
Data suggest that families will benefit from much more intensive support but this can prove difficult where diabetes teams are stretched already and alternative means of support do need to be developed (Brink et al, 2002). In the meantime, a conventional insulin regimen of twice-daily mixed insulin or an MDI regimen with long-acting background insulin and predetermined, fixed doses of rapid-acting insulin with meals, rather than carbohydrate counted adjusted doses, may be a suitable compromise for these families.
Insulin pump therapy
Insulin pump therapy has been approved by NICE for use in childhood (NICE, 2008). The NICE guideline states that a child under the age of 12 years will qualify for insulin pump therapy if MDI is considered impractical or inappropriate, which could include all children under the age of 12 years. Children over the age of 12 would need to fulfil adult criteria (NICE, 2008). In some areas, where insulin pump provision has been unproblematic, some services are starting children on insulin pumps at diagnosis, yet there are many centres where this is not possible for a number of reasons.
Data suggest that there is some benefit for insulin pump therapy in childhood mainly with respect to avoidance of hypoglycaemia (Pickup and Sutton, 2008). There are also emerging data to suggest that a combination of an insulin pump and continuous glucose monitoring (CGM) (worn continuously) can improve glycaemic control. The STAR 3 (Sensor-Augmented Pump Therapy for A1C Reduction) study examined the use of sensor-augmented insulin pump therapy. The results suggest that this can lead to improvements in HbA1c levels in both children and adults without an increased risk of hypoglycaemia (Bergenstal et al, 2010). CGM may be particularly useful in hypoglycaemia avoidance. A study of people with good glycaemic control – an HbA1c level of <7.0% (<53 mmol/mol) – showed that control could be improved even in these individuals and without hypoglycaemia (Juvenile Diabetes Research Foundation [JDRF] CGM Study Group, 2009).
In contrast, another study examining continuous rather than focused use of glucose sensors showed a benefit in adults over the age of 25 years but not in CYP, which may have been related to a lower rate of continuous use of the glucose sensor in younger people; CGM use for 6 or more days per week was 83% in those over 25 years of age, 30% in 15–24 year-olds and 50% of 8–14 year-olds (JDRF CGM Study Group et al, 2008).
There are some studies that have examined the benefits of self-monitoring of blood glucose (SMBG). Findings from these suggest that four blood glucose tests or more per day is associated with a 1 percentage point (13 mmol/mol) reduction in HbA1c (Levine et al, 2001). Yet in these studies it is difficult to ascertain whether it is those people who are most engaged with their diabetes who benefit the most from frequent blood glucose monitoring; doing six blood tests a day just to document them in a glucose diary is likely to be of limited therapeutic value. However, CYP on MDI regimens will be asked to perform four blood tests per day whereas CYP on pump therapy will be asked to perform a minimum of four to six tests per day.
Management of diabetes at school
The management of any child with a chronic illness in school can prove challenging. Dealing with blood tests and injections can prove even more so. With an increasing emphasis on intensive insulin regimens, schools are being asked to do more for children with diabetes, especially primary school children who are unable to do their own injections or, at the very least, need some adult supervision.
Although schools have a duty of care towards children with all sorts of medical needs, there is no legislation in the UK – or in many other countries globally – that specifies the supportive role of school staff (Diabetes UK, 2008). Studies of parents and children have shown that negative experiences are not uncommon in school. A number of children use the school toilets to do injections and blood glucose testing and there are concerns that school staff lack even a basic understanding of diabetes and so children rely on peers and not teaching staff for help (Newbould et al, 2007; Amillategui et al, 2009).
Data from Spain suggest that 44% of parents have had to rearrange their work schedules to support their children at school (Amillategui et al, 2007) and UK data show that if children were unable to administer their own insulin, 70% of schools expected a parent to come and give the injection (Diabetes UK, 2009). However, the difficulties for schools and teaching staff should not be underestimated and expectations for a high level of knowledge and understanding of diverse medical conditions from epilepsy, through congenital heart disease to diabetes, are significant.
Adolescence defines the period of time of transition from childhood to adulthood. It is a time of great change in biological, social and psychological terms. The impact of puberty on diabetes and the associated increased insulin requirement has already been discussed. If SMBG is performed regularly yet readings are often high due to the insulin resistance of puberty, this will be demoralising for adolescents. If high doses of insulin are given to overcome insulin resistance, frequent hypoglycaemia may ensue, which can be very embarrassing for adolescents, or could lead to unacceptable weight gain (especially in girls). It is not surprising that adolescents find it hard to keep up with the management of their diabetes. Box 2 highlights such issues.
In addition, having a chronic illness like diabetes at a time when fitting in with peers is of paramount importance is potentially very challenging. It is known that diabetes control deteriorates during this time – insulin doses may be omitted and clinic attendance can be erratic. A number of different approaches have been tried to improve adolescent engagement with diabetes management but there is no easy answer. Family support is crucial at this time and comprises “behaviours that foster in an individual feelings of comfort and belonging, and that he or she is basically accepted and approved of as a person by the parents and family” (Skinner et al, 2005).
Research shows a bell-shaped response to family support with both those getting too little and too much tending to do less well in terms of diabetes management (Gustafsson et al, 1987). Recurrent diabetic ketoacidosis is a significant risk especially in girls, those from lower socioeconomic backgrounds and those with pre-existing psychosocial issues (Chapman et al, 1988).
Risk taking is part of what may be considered normal adolescent behaviour – testing out the boundaries as a phase of personal development. These risks will have a variable impact on health but are likely to have a greater impact on health in those with an underlying chronic condition and may also be more prevalent in these younger people. In the UK, data suggest that 11% of 13- to 14-year-olds have smoked cigarettes with 4% smoking every day; 17% have been drunk at least once and 11% have tried some type of recreational drug (Office for Standards in Education, Children’s Services and Skills, 2008).
A large study from a multicentre collaboration across Germany and Austria showed that the prevalence of smoking in CYP with diabetes increased from 0.1% in under 11-year-olds to 28.4% in 15- to 20-year-olds (Hofer et al, 2009). Smokers had poorer glycaemic control (HbA1c 9.1% [76 mmol/mol] versus 8.0% [64 mmol/mol] in smokers versus non-smokers, respectively; P<0.0001) and there were also negative effects on diastolic blood pressure and lipid profile. However, this may reflect a group of individuals struggling to engage with their diabetes management in general. The long-term risks of micro- and macrovascular disease in people with type 1 diabetes are well described and so all adolescents should be informed of the dangers and offered smoking cessation support as and when available.
The UK has the highest rate of teenage pregnancies in Europe, five times the rate in The Netherlands and twice that of France (United Nations Statistics Division, 2010). Teenage pregnancies are associated with poorer health outcomes for both mother and child (Chen et al, 2007) but this may, in part, reflect the poorer socioeconomic status of the young women who become pregnant during adolescence. There do not appear to be any studies examining pregnancy rates and outcomes in adolescents with diabetes, although these do happen.
UK data from studies of adult women with diabetes are quite sobering. A study conducted by the Confidential Enquiry into Maternal and Child Health (CEMACH) found that only 41% of pregnancies were planned, 28% of women smoked (compared with 35% of the general maternity population) and only 27% took folic acid before pregnancy (CEMACH, 2007). The preterm delivery rate was 36% with a Caesarean section rate of 67% compared with 7% and 22%, respectively, in the general population. Women with diabetes also had a five-fold increased risk of stillbirth, a three-fold increased risk of perinatal mortality and a two-fold increased risk of fetal congenital anomaly.
Information and support for young girls and women with diabetes is vital to prevent unplanned pregnancies. Yet the correct methods and timing of providing such information to young girls remain unclear both for girls with diabetes and those in the general population (CEMACH, 2007). Adolescent girls will have a lot of exposure to healthcare – time that could be used for education with the cooperation and encouragement of parents or carers.
NICE guidance recommends that all children over the age of 12 years have an “annual review” with screening for microvascular complications (NICE, 2004). Young people need to have their blood pressure checked, retinal screening, urine screening for microalbuminuria and a blood test for assessment of renal function.
Up to 10% of people with type 1 diabetes will have autoimmune thyroiditis. If this develops during childhood then it can have a significant impact on cognitive development, and so NICE (2004) recommends that pre-pubertal children, <12 years old, have blood taken annually for renal function and thyroid function tests from diagnosis. Coeliac disease is also common in CYP with type 1 diabetes. In the author’s experience, this combination is extremely challenging – most of the foods recommended as part of a healthy diet, and which contain long-acting carbohydrate, contain gluten. “Unhealthy”, sugar-loaded food is often gluten-free.
Until recently, NICE (2004) recommended annual screening for coeliac antibodies in all CYP with diabetes regardless of age, and many asymptomatic people have been diagnosed and started on a gluten-free diet. The long-term health benefits of treating asymptomatic coeliac disease in CYP with diabetes are yet to be proven. More recently, NICE (2004) has decided that coeliac disease should be screened for at diagnosis only and individuals only re-tested if they develop symptoms. The psychological impact of being diagnosed with two lifelong, serious medical conditions that are almost incompatible does not appear to have been examined.
Childhood diabetes is a complex medical condition that requires healthcare professionals to have a large number of clinical, educational and psychological skills. There is no doubt that joint working with children and their families is as rewarding as it can be challenging. This is an area where joint working with primary care is still underdeveloped but could lead to improved outcomes for children and their families.
To complete the CPD module, click here.