Article Text


Endocrinology and diabetes

Statistics from


P. Paul, A. Ghatak, S. Kerr, C. Smith, M. Didi, J. C. Blair.Royal Liverpool Children’s hospital, Alder Hey, Liverpool, Liverpool, UK

Introduction: The incidence of cerebral oedema at presentation of diabetes mellitus is increased in children aged <2 years and those with severe diabetic ketoacidosis. Here we present a 10 year review of the metabolic profile of children aged <2 years at presentation of diabetes mellitus and report increased incidence of diabetic ketoacidosis and more profound metabolic decompensation than in older prepubertal children.

Patients and Methods: Blood pH, base deficit (BD), serum glucose (Glu), sodium (Na), osmolality, and urea at presentation of diabetes mellitus was studied in 21 children (11 female, age (mean (1 SD)) 1.39 (0.30) year) diagnosed between 1994 and 2003. Each child was matched to one or more pre-pubertal control subjects (25 male:19 female, age 8.05 (1.73) year) diagnosed with diabetes mellitus during the same week.

Results: 17/21 (84.8%) children aged <2 years presented in diabetic ketoacidosis compared with 10/44 (22.79%) children aged >2 years. Blood pH was significantly lower in children aged <2 years (7.17 (0.16)) than children aged >2 years (7.3 (0.14)), p 0.004 and BD was significantly higher (19.8 (7.6) mmol/l v 7.29 (11.4) mmol/l), p 0.003. There was no significant difference in serum Glu (30.05 (8.39) mmol/l v 28.43 (9.44) mmol/l), Na (135.3 (4.29) mmol/l v 132.74 (3.41) mmol/l), osmolality (307.96 (11.96) mmol/kg v 299.6 (12.04) mmol/kg), or urea (7.32 (3.41) mmol/l v 5.42 (1.88) mmol/l).

Conclusion: Diabetic ketoacidosis with severe metabolic decompensation is more frequent at presentation of diabetes mellitus in children aged <2 years. This may result from earlier ketogenesis reflecting lower glycogen stores in these young children and account for the excess of cerebral oedema observed in this age group.


R. Puttha1, D. Schapira2.1Chelsea and Westminster Hospital, London, UK; 2Royal Hampshire County Hospital, Winchester, UK

Background: The European Society of Paediatric Endocrinology and Diabetes UK recommend the use of insulin of 0.1 units/kg/hour in the first hour of diabetes ketoacidosis management, in children >5 years. Occasionally this dose of insulin results in a rapid drop in blood glucose levels. However, there is insufficient evidence whether lower doses of insulin could be used effectively and safely.

Aim: To study the effect of two different doses of insulin on the blood glucose values, acidosis and any clinical symptoms of cerebral oedema.

Method: We retrospectively reviewed the case notes of all children who presented with diabetic ketoacidosis, over a 6 year period between 1997 and 2002, to our hospital. In the initial hour, patients were administered insulin at either 0.05 unit/kg/hour as per our unit guideline (group 1) or 0.1 unit/kg/hour as per diabetes UK guideline (group 2), depending on individual physician’s choice. Twenty five episodes of diabetic ketoacidosis occurred in 14 children between 7 and 15 years. Ten of these patients were administered insulin at 0.1 unit/kg/hour and 11 were administered insulin at 0.05 unit/kg/hour. Four patients were excluded from the study as the data available were incomplete.

Results: For a similar degree of acidosis and mode of fluid administration, the drop in blood glucose value was higher in group 2 (range 10.2 to 22) compared with group 1 (range 2.4 to 9.4 millimoles per decilitre). Data regarding the change in acidosis were available for four patients and was less than 0.3 units in all four patients (three from group 2). One patient had signs suggestive of raised intracranial pressure. Three patients complained of mild headache. All four patients were from group 2. No child in group 1 complained of headache or elicited features of raised intracranial pressure.

Conclusion: Rapid reduction of blood glucose levels can contribute to a significant drop in serum osmolality. Use of lower dose of insulin might be safer, especially in patients with pH of less than 7.1 and when fluid boluses are needed for dehydration. Our study is limited by small numbers. Further studies on this topic might help resolve this issue.


M. A. Riddle, N. C. Trevelyan, S. Cottrell, P. R. Betts.Southampton General Hospital, Southampton, UK

Aim: The recent advent of insulin analogues has encouraged the use of flexible and innovative multiple insulin regimens in many centres, using four or more injections per day (basal bolus regimen). Over the past 2 years in our paediatric diabetic clinic we have encouraged patients to change from standard twice daily insulin to a basal bolus regimen. We undertook a retrospective case notes review to look at the impact of this change on: HbA1C; body mass index standard deviation scores (BMI SDS); incidence of acute complications (significant hypoglycaemia, diabetic ketoacidosis); and perceptions of wellbeing and quality of life scores for patients and their families.

Method: All children who had changed from a twice daily insulin injection regimen for ⩾1 year were identified. A retrospective analysis of patient files, annual review summaries, and hospital pathology computer database was used to record three monthly HbA1C and BMI in the 1 year period pre and post regimen change for each patient. Number of episodes of diabetic ketoacidosis and significant hypoglycaemia (requiring hypostop) in this period was recorded. A questionnaire was sent to those patients who had changed insulin regimen. This was designed to assess perceptions of glycaemic control, wellbeing, satisfaction, and problems associated with the change in regimen.

Results: Forty eight children identified who had changed from two to four injections per day (20 male, 28 female). Mean age at time of change was 12.8 years (range 3.83 to 17.08 years). Mean HbA1C year prior to change was 9.2%, and year post change 8.6% (p 0.015; 95% CI 0.2 to 0.8). Mean BMI SDS prior to change was 0.805 (range −1.03 to 2.95), and 1 year post change 0.968 (range, −0.68 to 2.94) (p 0.347, 95% CI −0.5 to 0.18). 8/48 (17%) had ⩾one significant hypoglycaemic episodes while on two injections, of these children none reported hypoglycaemic episodes during the year post change. 2/48 (4%) had hypoglycaemic episodes in year post change, neither reported hypoglycaemic episodes in the year pre change. 3/48(6%) had one or more diabetic ketoacidosis admissions pre, and no children had episodes of diabetic ketoacidosis in the year post change. Of the children changing insulin regimen 70% thought that this had improved their diabetes control, and 80% had an improved sense of wellbeing after changing.

Conclusion: The use of a basal bolus insulin regimen has resulted in improved glycaemic control, and no increase in BMI SDS in our cohort. There was also a reduction in both hypoglycaemic episodes and diabetic ketoacidosis in the year following change of insulin. Our patients reported improvement in quality of life and wellbeing on the new regimen.


C. Jackson1, J. Richer1, J. A. Edge2.1Department of Clinical Psychology, John Radcliffe Hospital, Oxford, UK; 2Department of Paediatrics, John Radcliffe Hospital, Oxford, UK

Aim: Siblings of children with chronic conditions have been shown to be at higher risk of emotional and behavioural problems, although some research has demonstrated positive effects of having a sibling with a chronic condition. Previous research has investigated risk and resilience factors associated with the adjustment of children with chronic conditions and their parents. The aims of this study were to investigate sibling adjustment to type 1 diabetes mellitus, and to look at risk and resilience factors in the siblings and the family, using a cross sectional questionnaire survey design.

Method: Forty one families were recruited from a children’s diabetes clinic. All families on the clinic database were contacted and of these, 21% met the inclusion criteria (a sibling between ages 7 and 16 years) and agreed to participate, and 21% who were eligible did not respond. From each family, one parent and one sibling of the child with type 1 diabetes mellitus participated. Parents completed questionnaires measuring sibling adjustment, and also measures of major life events, social support, and parenting stress, to collect data on likely predictors of sibling adjustment. Siblings completed questionnaires assessing cognitive appraisals and coping strategies. A semi-structured interview was also administered to siblings.

Results: Siblings were found to be better adjusted than the normative data obtained from a national mental health survey.1 Factors associated with poorer sibling adjustment were: higher sibling age at diagnosis, higher levels of parenting stress, more difficult sibling temperament, poorer adjustment of the child with type 1 diabetes mellitus, higher levels of parental distress, and more negative sibling perceptions of diabetes and its impact on the family. Results suggest that sibling perceptions of diabetes and parental distress are important predictors of sibling adjustment to type 1 diabetes mellitus.

Conclusion: The findings from this study emphasise the relationships between the adjustment of the sibling, and that of the child with type 1 diabetes mellitus and their parents. Many parents worry about how the siblings may cope with the diabetes, but the results of this study are generally reassuring. It may be helpful to provide age appropriate information for siblings to encourage an accurate perception of diabetes.



F. Regan, A. Sleigh, T. Saukkonen, R. Williams, D. Dunger.University of Cambridge, Cambridge, UK

Background: The hyperinsulinaemic, euglycaemic clamp is the gold standard assessment for insulin sensitivity. Magnetic resonance spectroscopy quantification of intramyocellular lipid concentration (IMCL) may be a faster and non-invasive technique to determine peripheral insulin sensitivity, as intramyocellular lipids may decrease insulin sensitivity by inhibiting muscle glucose uptake. Previous studies have demonstrated an inverse relationship between IMCL concentration and total body glucose utilisation (or m-value) in a hyperinsulinaemic euglycaemic clamp. People with type 1 diabetes are known to be insulin resistant but there have been few studies to date documenting their IMCL.

Aims: To assess intramyocellular lipids and insulin resistance in patients with type 1 diabetes and compare values with normal healthy volunteers.

Methods: Ten adolescents and young adults with type 1 diabetes and 27 healthy controls were investigated. Insulin sensitivity was measured using a standard hyperinsulinaemic, euglycaemic clamp over a 2 hour period following an overnight fast. Intramyocellular lipids were assessed in the soleus muscle using 1H spectrum measured with a 3.0 T magnet NMR spectrometer.

Results: Diabetic subjects were more insulin resistant as measured by the clamp (that is, lower m value) and had higher IMCL, despite similar BMI. After adjustment for BMI, greater insulin resistance was associated with increased IMCL (r 0.43, p 0.008).

Conclusion: Type 1 diabetic subjects are insulin resistant when compared with healthy controls. Not only is their total body insulin sensitivity reduced but also their peripheral insulin sensitivity as measured by IMCL. This study confirms preliminary findings that IMCL concentration is negatively correlated with insulin sensitivity, as measured by the clamp, indicating it is a valid, quick, and non-invasive measure of peripheral insulin sensitivity.


C. P. H. Ibrahim, N. J. Wild, W. Scmidt.North Cheshire Hospitals NHS Trust, Warrington, UK

Aims: To estimate the prevalence of obesity and overweight in a defined population of children with insulin dependent diabetes mellitus and to compare the prevalence with available national data in similar age groups.

Methods: All children with insulin dependent diabetes mellitus under the care of the local paediatric diabetic team were considered for inclusion. Children with cystic fibrosis were excluded. International obesity task force criteria based on body mass index were used for classification of obesity or overweight. National data for comparison were derived from the Health Survey of England, 2002.

Results: 131 children with insulin dependent diabetes mellitus were studied (74 boys, 57 girls). The median age of males was higher (13.9 v 11.24; p 0.016). The median dose of insulin was marginally higher in males (1 U/kg v 0.98 U/kg; p 0.05). There was a higher proportion of boys who were overweight (35.1% v 25.05%). There was no correlation of overweight or obesity with age, duration of diagnosis, or insulin dose. The proportion of overweight children was significantly higher in children with HbA1c <8.5% (p 0.032). Compared with the national prevalence data, prevalence of overweight was significantly higher among males (see table 1). The prevalence among girls was similar to the national figures (see table 2). The trend was similar when analysed separately in the adolescent age group. The median HbA1c was similar in both sexes (8.4 v 8.3; p 0.93).

Abstract G37

Abstract G38 Table 1 Proportion of obese and overweight diabetic boys (2–18 years)

Abstracts G38 Table 2 Proportion of obese and overweight diabetic girls (2–18 years)

Abstracts G39

Abstract G41

Conclusions: It is difficult to achieve a good balance between diabetic control and body weight in children with insulin dependent diabetes mellitus. The prevalence of overweight among children with insulin dependent diabetes mellitus, especially boys, is significantly higher than that in the general population. The higher prevalence of overweight in boys compared with girls could not be explained by better glycaemic control. Children with insulin dependent diabetes mellitus require a sustained lifestyle educational programme to prevent future morbidity related to obesity and overweight.


M. L. Yeap2, J. Cornish2, A. Oakhill1, C. P. Burren2, F. J. Ryan3, E. C. Crowne1.1Bristol Children’s Hospital, Bristol, UK; 2Bristol Children’s Hospital, Bristol, UK; 3John Radcliffe Hospital Oxford, UK

Aims: To determine if prepubertal survivors of leukaemia have decreased bone mineralisation post bone marrow transplant or standard chemotherapy treatment compared with age matched controls.

Methods: Eighty patients aged 5–12 years were recruited, group 1 (bone marrow transplant with total body irradiation TBI)) n = 23, 9 female:14 male; group 2 (leukaemia with no cranial irradiation) n = 26, 12 female:14 male; and group 3 (controls) n = 31, 12 female:19 male. Group 1 and group 2 patients were treated on standard chemotherapy and transplant protocols and were >1 year off treatment. Patients had standard auxology and Lunar Prodigy DEXA Scanner was used to take a single measurement of total body bone mass, lean body mass, and also femoral neck and lumbar L2-4 bone mass. Mean time from diagnosis is group 1 6.2 years and group 2 6.9 years.

Results: See table. In view of short stature in group 1 (decreased height SDS), lean body mass and bone mineral density was corrected for height and expressed as lean body mass/height SDS, fat free mass index, and percentage bone mineral content, respectively. Mean total bone mineral content/area, bone mineral content/height, and bone area/height did not show any significant differences. Volumetric bone mineral density of the lumbar L2-4 spine was not significantly decreased between the 3 groups (group 1: 0.31; group 2: 0.29; and group 3: 0.32). Volumetric bone mineral density of the femoral neck in group 1: 0.27, group 2: 0.31, and group 3: 0.32. With no significant differences between the 3 groups or between males and females.

Conclusion: Prepubertal survivors of leukaemia treated by bone marrow transplantation have decreased total body bone mineral content, which may be due to decreased total lean body mass as lean tissue mass is one of the factors that has an impact on bone mineralisation. Prepubertal chemotherapy treated survivors have normal lean tissue mass, total and regional bone mineralisation.


B. W. M. Van Bon, D. S. Gardner, J. Dandrea, T. Stephenson, M. E. Symonds.Centre for Reproduction and Early Life, University of Nottingham, Nottingham, UK

Introduction: Maternal nutritional restriction at different stages of gestation has been proposed to be one factor determining predisposition to later diabetes. The precise timing in pregnancy of this adverse outcome remains to be established. The aim of the present study was therefore to determine the effect of nutritional restriction in utero on glucose tolerance and insulin secretion in the resulting offspring as young adults.

Methods: Fourteen singleton bearing sheep were entered into the study. Five were nutrient restricted over the first (NRE) or last (NRL) 30 days of gestation when they were fed 50% of their total calculated metabolisable energy requirements for body weight and pregnancy. At all other stages of gestation they were fed to 100% of requirements, together with a control group (n = 5). The offspring were then raised to 1 year of age when they all underwent a GTT following 24 h fasting. Glucose was administered (0.5 g/kg intravenously in 0.9% NaCl) and repeated blood samples were taken over a 2 hour period. These were then assayed for plasma glucose and insulin (RIA) concentrations. Significant differences between groups were assessed by ANOVA.

Results: There was no effect of diet on the maximal change in plasma glucose, insulin, or time to restore plasma glucose to baseline. The time taken for plasma insulin to return to baseline after the GTT was significantly increased in NRL group, which also showed decreased glucose tolerance (control 1111 (SD 054); NRE 1340 (206); NRL 1619 (170) min.mmol.L−1 (p<0.05)) despite significantly increased absolute insulin secretion (control 70 (9); NRE 53 (11); NRL 116 (22) min.nmol.L−1 (p<0.05)). NRL also had ∼2-fold more adipose than the other groups.

Conclusions: The timing of maternal nutrient restriction through pregnancy is a major factor determining long term insulin and glucose sensitivity which may be mediated in part by changes in the handling of glucose by adipose tissue.


S. Kanumakala1, R. Greaves2, C. Pedreira2, S. Donath2, G. Warne2, M. Zacharin2, M. Harris3.1Brighton & Sussex University Hospitals NHS Trust, Brighton, UK; 2Royal Children’s Hospital, Melbourne, Australia; 3Mater Children’s Hospital, Brisbane, Australia

Background: Morbid obesity is common after hypothalamic damage, which is often resistant to standard therapeutic interventions. Hypothalamic dysfunction is also thought to underlie the obesity that is typical of Prader Willi syndrome. Elevated fasting levels of appetite stimulating hormone, ghrelin, have been described in Prader Willi syndrome. The aim of this study was to determine whether fasting ghrelin levels are increased in children with hypothalamic obesity.

Methods: Fasting total ghrelin, insulin, glucose, and leptin were measured in children with hypothalamic obesity (n = 16) and compared with age and sex matched normal obese children (n = 16) and non-obese children (n = 16). The two obese groups were also matched in terms of body mass index, although children with hypothalamic obesity were shorter as compared with their obese controls.

Results: Obese children, as a whole, had lower fasting total ghrelin levels as compared with normal controls (p 0.01). However, there was no difference between the fasting total ghrelin levels between children with hypothalamic obesity and normal controls (p 0.88). Similarly, there was no difference between fasting insulin and leptin levels between the two obese groups. Full results are reproduced in the table with data presented as medians (inter quartile range) and inter-group comparisons were analysed using the Kruskall-Wallis or Mann Whitney test.

Conclusions: These data suggest it is unlikely that an elevation in fasting ghrelin is responsible for the obesity that occurs after hypothalamic damage. Therapeutic interventions aimed at reducing fasting ghrelin may prove ineffective in controlling weight gain in this group.


T. Y. Segal, T. Hernandez-Crespo, P. C. Hindmarsh, R. M. Viner.Royal Free & University College Medical School, London, UK

Aims: Despite the rising prevalence of obesity in childhood, there is little consensus on whether pharmacological interventions are justified in children. On the basis of brief randomised trials, metformin is being increasingly used in those with abnormal glucose/insulin homeostasis. However, there are little data on the long term outcomes of, or predictors of response to, metformin in clinical situations.

Methods: We prospectively identified 157 obese (body mass index (BMI) ⩾95th centile) children and adolescents undergoing an oral glucose tolerance test (OGTT; 1.75 mg of glucose/kg) since 2001, of whom 74 (47%) continued to have clinical follow up. All subjects subsequently received standard dietetic education. Metformin (low dose <1 g/day or high dose 2 g/day) was used in those with fasting hyperinsulinaemia and/or hyperinsulinaemia during the OGTT. Hyperinsulinism was defined by pubertal stage: prepubertal ⩾15 mU/l; mid-puberty ⩾30 mU/l; and post-puberty ⩾20 mU/l. Responders to metformin were defined a priori as those with reduction in BMI ⩾0.1 SD.

Results: Twenty six (35%; 21 females) subjects had been treated with metformin (mean age 13.2 year, range 4.6–18 year). Controls: n = 48; 37 females; mean age 11.0 year. Mean BMI z score at baseline was not significantly different between groups. Those in the metformin group had significantly lower BMI at 12 months, falling 0.4 SD (SE 0.2) compared with a fall of 0.2SD (SE 0.05) in controls (F = 7.3; p 0.01). Response to metformin was greatest in those with fasting hyperinsulinism, with mean BMI reduction of 0.4 SD compared with 0.07 SD in those in the metformin group with normal fasting insulin. Sex, age, pubertal stage, ethnicity, presence of acanthosis nigricans, BMI before treatment, and metformin dose were not associated with response.

Conclusions: Benefits of metformin in reduction of BMI in obese non-diabetic children and adolescents are maintained at 12 months. Those with fasting hyperinsulinaemia appear more likely to benefit than those with post-prandial hyperinsulinaemia.

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.