Kharen L. Clayton

Growth During Childhood and Final Height in Type 1 Diabetes

The growth of 184 children with Type 1 diabetes was analysed using data collected prospectively in the Oxford district between 1969 and 1992. The overall mean height standard deviation score (Ht SDS ± SD) at diagnosis was 0.35 ± 1.05 which was significantly greater than the national standard of Tanner (1966). However, there is evidence of a secular trend in the heights of Oxford children over the last 20 years when compared with Tanner. When data from children with diabetes were compared with local controls, it was only the children aged 5–10 years at diagnosis who were taller (Ht SDS ± SD, 0.58 ± 1.14, versus 0.31 ± 0.90, n = 73, p ± 0.05). Those diagnosed under the age of 5 years (n = 37) were shorter (Ht SDS 0.12 ± 0.93) and those diagnosed aged more than 10 years (n = 74) were similar in size (Ht SDS 0.22 ± 0.98) to controls. These differences could not be explained by social class. Loss of height occurred between diagnosis and puberty, particularly in those diagnosed between the ages of 5 and 10 years. The pubertal growth spurt was blunted in all groups but this abnormality was more profound in the girls (mean peak height velocity SDS ‐1.09 ± 1.02, p ± 0.0005) than in the boys (mean peak height velocity SDS ‐0.5 ± 1.14, p ± 0.025). The mean final height SDS was ‐0.74 ± 0.96 in those diagnosed < 5 years, 0.00 ± 1.26 in those diagnosed between the ages of 5 and 10 years and 0.09 ± 1.10 in those aged more than 10 years at diagnosis. Overall final heights were not significantly different from the mid‐parental height SDS. It is of interest that those children who were diagnosed under 5 years had short parents (mid‐parental height SDS, ‐0.39 ± 1.09). Height at diagnosis may reflect the heterogeneity of genetic and environment factors in the aetiology of diabetes. Growth following diagnosis is abnormal although there is no major impact on final height.

The effects of repeated daily recombinant human insulin-like growth factor I administration in adolescents with type 1 diabetes

Reduced insulin‐like growth factor bioactivity has been linked to poor metabolic control and growth hormone hypersecretion in adolescents with Type 1 diabetes. The safety and efficacy of recombinant human insulin‐like growth factor I administered subcutaneously in a dose of 40 μg kg−1 for 28 days was studied in a group of 6 adolescent male subjects with Type 1 diabetes (aged 13.6–19.4 years, puberty stage 3–5). After a 4‐week run‐in period (week 4 to day 0) recombinant human insulin‐like growth factor I was administered for 4 weeks (day 0 to week +4) before a run‐out of a further 4 weeks duration (week +4 to +8). HbA1c levels were measured throughout the study and overnight profiles were undertaken to study levels of insulin‐like growth factor I, insulin‐like growth factor binding protein‐3, and growth hormone concentrations (week —1, day 0, and week +4). The injections were well tolerated and hypoglycaemia was not problematic at any stage of the study. Recombinant insulin‐like growth factor I administration appeared to lead to a sustained increase in insulin‐like growth factor I levels (week —1; 198 ± 16 ng ml−1, week +4; 422 ± 18 ng ml−1, mean ± SEM; p = 0.03). Insulin‐like growth factor binding protein‐3 concentrations (n = 6) increased in 5 subjects (week —1; 4.5 ± 0.3 μg ml−1, week +4; 5.1 ± 0.4 μg ml−1) and mean overnight growth hormone decreased (week −1; 14.0 ± 3.1 mU I−1, week +4; 7.6 ± 1.7 mU I−1) during the period of study but these differences were not statistically significant. HbA1c levels fell significantly at the time of rhIGF‐I administration (day 0; 10.4 ± 1.9% vs week +4; 9.4 ±1.9%; p = 0.03) despite a reduction in subcutaneous isophane insulin dose from 0.50 ± 0.02 U kg−1 to 0.41 ± 0.02 U kg−1 (p = 0.03). There was no significant change in biochemical and haematological indices, glomerular filtration rate or urinary albumin excretion. The restoration of IGF‐I levels in adolescents with Type 1 diabetes may have a beneficial impact on glycaemic control.

The effects of recombinant human insulin-like growth factor I on growth hormone secretion in adolescents with insulin dependent diabetes mellitus.

OBJECTIVE It has been proposed that low IGF‐I levels and reduced IGF‐I bioactivity may lead to elevated GH levels in adolescents with insulin dependent diabetes (IDDM). We have therefore studied the effects of human recombinant insulin‐like growth factor I (rhIGF‐I) administration on GH levels and GH secretion in adolescents with IDDM.

The relationship between overnight GH levels and insulin concentrations in adolescents with insulin-dependent diabetes mellitus (IDDM) and the impact of recombinant human insulin-like growth factor I (rhIGF-I).

OBJECTIVE We examined the relationship between GH concentrations and free insulin concentrations, used as an index of insulin sensitivity, before and after recombinant human insulin‐like growth factor I (rhIGF‐I) administration in adolescents with insulin‐dependent diabetes mellitus (IDDM).

Serum insulin-like growth factor II levels in normal adolescents and those with insulin dependent diabetes mellitus

OBJECTIVE Unlike IGF‐I and its principal binding proteins, data regarding IGF‐II levels have not been well defined in normal subjects and those with insulin‐dependent diabetes mellitus (IDDM). We have therefore measured IGF‐II, as well as IGF‐I, and IGFBP‐3, levels In a large cohort of subjects with IDDM and in age/sex matched controls.

THE CHARACTERISTICS OF GROWTH HORMONE (GH) PULSATILITY IN NEONATES

GH is now recognised to be important in early postnatal growth, High GH levels nave been documented but the pulsatility has not been defined because of difficulties in obtaining frequent blood samples. We have developed an automated computerised technique which enables microsamples of blood to be withdrawn through an intravenous (IV) cannula. Blood samples of 40ul were collected in a validated dilution of 1:4, with heparinlsed saline. GH was measured by immunoradiometric assay. The sensitivity was 0.2mU/l and the intra-assay CV[%] was 9.6, 7.7, and 4.4 at concentrations of 1.0, 5.0 and 25mU/l respectively. GH profiles (10m sampling over 12hr) from 5 normal babies at the end of IV drug therapy were analysed by pulsar and time series analysis.The dominant pulse periodicity [Fourier Transform) in these babies was 90-100 minutes.We have demonstrated high mean and baseline GH levels and a fast frequency GH pulsatility in newborn babies.

THE RELATIONSHIP BETWEEN GH, IGF-1, IGFBP-3 AND GH BINDING PROTEIN (GHBP) IN NORMAL AND ADOLESCENTS WITH INSULIN-DEPENDENT DIABETES MELUTUS (IDDM)

The relationship between GH1 IGF-I, GHBP and IGFBP-3 level in normals and adolescents with IDDM have not been adequately defined. We have measured IGF-I and IGFBP-3(RIA)), GHBP(LIFA) in samples from 89 normal (40F, 49M) and 94 IDDM (54F, 40M) subjects matched for puberty stage (G1-5, B1-5). Coincident mean overnight GH data (15 minute sampling, 20.00h-08.00h, IRMA) were available from 33 normal and 30 IDDM subjects. Maximal IGF-I levels were noted at G4, B3 in normals and G5, B3 in IDDM. Levels were significantly lower in diabetes at 65 (p<0.05) and B4 (P<0.002). In normals and diabetics maximal IGFBP-3 were noted at B3 (6.7±1,2 mg/ml and 5.3±1.0 mg/ml respectively) and 64 (6.4±0.7 mg/ml and 4.6±1.0 mg/ml) with reduced levels in the diabetics at B2, B4 and G2-5. GHBP did not change significantly during puberty in normals or IDDM but the latter had lower levels at B3-41 G1-3. GH concentrations were greatest at B2-3 and G4-5 in nonals and were greater in diabetics at all stages. In normals there was a correlation after log transfonation between GH and GHBP (r=−0.59, p<0.001) and IGF-I and IGFBP-3 (r=0.51, p<0.001). In IDDM the relationship between IGF-I and IBFBP-3 (r=0.55, p<0.001) was retained, there was no correlation between GH and GHBP1 however a weak correlation between GH and IGF-I (r=0.38, p<0.04) was evident. During puberty GH is increased, yet IGF-I, IGFBP-3 and GHBP are reduced in IDDM. The normal negative correlation between GH and GHBP is lost. The relationship between IGF-I and IGFBP-3 is similar in both groups of subjects.