Low risk HLA-DQ and increased body mass index in newly diagnosed type 1 diabetes children in the Better Diabetes Diagnosis study in Sweden.

OBJECTIVE: Type 1 diabetes and obesity has increased in childhood. We therefore tested the hypothesis that type 1 diabetes human leukocyte antigen DQ (HLA-DQ) risk genotypes may be associated with increased body mass index (BMI). DESIGN: The type 1 diabetes high-risk HLA-DQ A1*05:01-B1*02:01/A1*03:01-B1*03:02 genotype along with lower risk DQ genotypes were determined at the time of clinical onset by PCR and hybridization with allele-specific probes. BMI was determined after diabetes was stabilized. SUBJECTS: A total of 2403 incident type 1 diabetes children below 18 years of age were ascertained in the Swedish national Better Diabetes Diagnosis (BDD) study between May 2005 to September 2009. All children classified with type 1 diabetes, including positivity for at least one islet autoantibody, were investigated. RESULTS: Overall, type 1 diabetes HLA-DQ risk was negatively associated with BMI (P<0.0008). The proportion of the highest risk A1*05:01-B1*02:01/A1*03:01-B1)03:02 genotype decreased with increasing BMI (P<0.0004). However, lower risk type 1 diabetes DQ genotypes were associated with an increased proportion of patients who were overweight or obese (P<0.0001). Indeed, the proportion of patients with the low-risk A1*05:01-B1*02:01/A1*05:01-B1*02:01 genotype increased with increasing BMI (P<0.003). The magnitude of association on the multiplicative scale between the A1*05:01-B1*02:01/A1*05:01-B1*02:01 genotype and increased BMI was significant (P<0.006). The odds ratio in patients with this genotype of being obese was 1.80 (95% confidence interval 1.21-2.61; P<0.006). The increased proportion of overweight type 1 diabetes children with the A1*05:01-B1*02:01 haplotype was most pronounced in children diagnosed between 5 and 9 years of age. CONCLUSIONS: Susceptibility for childhood type 1 diabetes was unexpectedly found to be associated with the A1*05:01-B1*02:01/A1*05:01-B1*02:01 genotype and an increased BMI. These results support the hypothesis that overweight may contribute to the risk of type 1 diabetes in children positive for HLA-DQ A1*05:01-B1*02:01.

Determinants of growth in diabetic pubertal subjects.

OBJECTIVE: To analyze the relationship among metabolic control, IGF-I, and growth in pubertal diabetic subjects. RESEARCH DESIGN AND METHODS: In 72 diabetic children, we have studied the pattern of change of IGF-I, IGF-I SD score, IGF binding protein (BP)-1, and growth rate in different pubertal stages and have analyzed their relation to age sex, weight/length index, HbA1c, insulin concentration, insulin dose, and dehydroepiandrosteronesulfate (DHEAS). RESULTS: The serum IGF-I values increased up to Tanner stage 4 and thereafter decreased, whereas IGFBP-1 showed the inverse pattern. When transforming the IGF-I values into SD scores, correcting for age, sex, and pubertal stage, it was shown that the deviation from normal values increased with increasing pubertal stage in boys, but was equal in stages 3-5 in girls. Using multiple regression analysis, HbA1c, insulin dose, and DHEAS were significantly correlated to IGF-I SD score (R2 = 0.253, P = 0.001). IGFBP-I levels in the afternoon were within normal range. LogIGFBP-1 showed an inverse correlation, to insulin concentration in single correlation (r = -0.26, P = 0.02). In single correlation, growth rate correlated significantly to insulin dose (r = 0.25, P = 0.03). In a multiple regression analysis, only DHEAS and IGF-I SD score were found to be significantly correlated to growth rate (R2 = 0.370, P < 0.001). The 18 adolescents who had reached their final height did not deviate from their target final height, according to their recorded growth since birth. CONCLUSIONS: In a group of fairly well-controlled diabetic children, the normal increase in IGF-I during puberty is blunted. Despite decreased IGF-I levels, target final height was attained, probably because of adequate insulin compensation leading to normal IGFBP-l, thus adequate bioavailability of IGF-I. Our results point out the importance of sufficient exogenous insulin in the period of rapid linear growth.

Increased 24 h mean insulin-like growth factor binding protein-3 proteolytic activity in pubertal type 1 diabetic boys.

Hyperglycaemia and increased variability of blood glucose in pubertal children with type 1 diabetes may be related to increased growth hormone (GH) secretion and insulin resistance. The role of changes in insulin-like growth factor-I (IGF-I) bioavailability for the glycaemic control in these patients has not been completely elucidated. In particular, the possible role of increased IGF binding protein-3 (IGFBP-3) proteolysis reported in other insulin resistant states awaits further characterization. The aims of this study were to assess if hyperglycaemia in children with type 1 diabetes was associated with changes in free dissociable IGF-I (fdIGF-I) and IGF binding protein-3 protease activity (IGFBP-3-PA) and if increased insulin resistance during puberty was associated with changes in IGFBP-3-PA in healthy and diabetic children. In diabetic boys in the period of maximal linear growth (Tanner stage 3, n = 5), the mean level and the variability of IGFBP-3-PA, determined every second hour throughout 24 h, were significantly higher both compared to postpubertal diabetic boys (n = 6; P = 0.003 and P = 0.001, respectively), and to age matched healthy boys (n = 4; P = 0.006 and P < 0.001 respectively). This activation of IGFBP-3-PA was most prominent during the day time. The mean 24 h blood glucose level (determined hourly) was the only parameter studied that significantly predicted the changes in mean 24 h IGFBP-3-PA in the diabetes group. The mean 24 h concentrations of fdIGF-I were decreased in the diabetic boys compared to the healthy controls but statistical significance was only achieved in Tanner Stage 5 (p = 0.03). We speculate that the elevated levels of IGFBP-3-PA in Tanner 3 diabetic boys are related to deteriorated glucose homeostasis and that it may be a compensatory mechanism to attenuate the decrease in fdIGF-I in order to partly restore insulin sensitivity and glycemic control.

Determinants of blood glucose variability in adolescents with insulin-dependent diabetes mellitus.

Metabolic control and blood glucose variability in children with insulin-dependent diabetes mellitus (IDDM) during and after puberty were studied. Seventy-two children (43M, 29F), aged 10-19 years, with a 2-16-year duration of IDDM participated in the study. Fourteen of the patients were prepubertal (Tanner stage 1), 27 pubertal (Tanner 2-4) and 31 postpubertal (Tanner 5). They performed self-monitoring of blood glucose (SMBG) five times daily, every 2 days for 4 weeks. The SD (SDbg) for all values in each patient was calculated as a measure of blood glucose variability. Weight-length index, linear growth velocity and Tanner stage were recorded. Hemoglobin (Hb)A1c, alkaline phosphatase and sex hormone levels in serum were analyzed. Subjectively experienced hypoglycemic episodes were recorded. HbA1c levels showed no relation to Tanner stage. SDbg was lower in stage 5 than in stages 2-4 (p = 0.02). There was no significant correlation between HbA1c and SDbg, but the variability was significantly lower in individuals with mean blood glucose in the lower quartile compared with those in the upper three quartiles (p < 0.001). Alkaline phosphatase concentration, as a measure of growth velocity, was the main independent determinant of SDbg (r = 0.35, p < 0.005). There was an inverse correlation between levels of sex hormones and SDbg. We conclude that blood glucose variability is lower after than during puberty. This variability seems to be related to linear growth velocity or its biochemical marker.

Insulin-like growth factor binding protein-1 as glucose regulator in adolescent boys with type 1 diabetes.

UNLABELLED: The aim of the present study was to investigate whether the diurnal variability of B-Glucose is dependent on GH, IGF-I and IGFBP-1 levels, apart from insulin, and if there is any difference between Tanner stages 3 and 5. Five boys in Tanner stage 3 and 6 boys in stage 5 with type 1 diabetes were included. Blood was continuously collected from a cubital vein for 24 h. S-Insulin, S-GH, S-IGF-I and S-IGFBP-1 were analysed. B-Glucose was analysed hourly at bedside. One week before and 1 wk after the 24-h study period the participants performed self-monitoring of blood glucose (SMBG) during normal physiologic conditions. In the 24-h profile of B-Glucose, insulin, IGFBP-1 and GH, we found a significant positive correlation between B-Glucose and log IGFBP-1 (r = 0.5, p = 0.005) and an inverse correlation to insulin (r = -0.5, p = 0.004) but no correlation to logGH (r = -0.04, p = 0.831). In multiple regression analysis, B-Glucose was still significantly correlated to log IGFBP-1, when adjusting for insulin and GH, in Tanner stage 5. We found a difference between Tanner stages 3 and 5 in the variability of B-Glucose over a longer period during normal daily activity (p = 0.02), but not over the 24-h study period. CONCLUSION: We have demonstrated in type 1 diabetes adolescent boys a relationship between simultaneously measured blood-glucose and IGFBP-1 levels independent of the insulin and GH levels, suggesting that the free fraction of IGF-I influences the glucose metabolism.