The prevalence and clinical characteristics of celiac disease in juvenile diabetes in Wisconsin.

BACKGROUND: The relationship between celiac disease and juvenile diabetes has long been known. Only a single study in the United States, from Buffalo, New York, has reported the prevalence of celiac disease in a pediatric diabetic population. This study was conducted to determine the prevalence and clinical presentation of celiac disease in children and adolescents with juvenile diabetes in Wisconsin, USA, using serum antiendomysial antibody as a screening test. METHODS: Two hundred eighteen patients with diabetes (113 males; age range, 4-21 years) and 117 age-and gender-matched control participants were tested for immunoglobulin A endomysial antibody. Patients with positive results were offered a small bowel biopsy. A questionnaire regarding abdominal pain, diarrhea, and growth failure was completed by the parents. RESULTS: Seventeen of 218 diabetic patients (7.7%) had positive endomysial antibody. All control participants had negative results for the endomysial antibody. Small bowel biopsy was performed in 14 patients. Ten patients had villous atrophy. In one patient without villous atrophy, a repeat biopsy 2 years later showed villous atrophy, and two patients had increased intraepithelial lymphocytes without villous atrophy. Seventy percent of the patients with celiac disease were asymptomatic. The reported symptoms were abdominal pain and diarrhea (n = 1) and growth failure (n = 2). Two patients with celiac disease had Down syndrome. CONCLUSIONS: The prevalence of celiac disease in children with juvenile diabetes in Wisconsin is at least 4.6%, which is comparable with European and Canadian studies. Because patients without villous atrophy may have latent celiac disease, the prevalence may be even higher. All children with juvenile diabetes should be screened for celiac disease.

A study of chromosomal aberrations and chromosomal fragility after recombinant growth hormone treatment.

Increased chromosomal rearrangements and chromosomal fragility have been previously observed in lymphocytes of children treated with human GH, implying that treatment could predispose to malignancy. Twenty-four children with classic GH deficiency, neurosecretory GH dysfunction, and Turner syndrome were treated with recombinant human GH (0.3 mg x kg(-1) x wk(-1)). Metaphase cells were assessed for spontaneous chromosomal and chromatid aberrations at baseline and 6 mo into treatment. There were no significant differences in aberrations between baseline and the 6-mo samples. However, the mean frequency of chromatid-type aberrations on a per cell basis was significantly higher than at baseline, 0.0088 versus 0.0064 aberrations per cell (p < 0.024). Two patients contributed inordinately to this increase. A third sample from these two patients was almost identical to their baseline samples. Cells were also irradiated in vitro (3 Gy) to assess chromosomal fragility. After irradiation, no patient showed a significant difference for any aberration type, although there was a significantly lower frequency of ring chromosomes on a per cell basis in the 6-mo samples (p < 0.001). We find no evidence that GH therapy influences spontaneous chromosomal aberrations or chromosomal fragility.

Changes in thyroid hormone levels during growth hormone therapy in initially euthyroid patients: lack of need for thyroxine supplementation.

The occurrence of central hypothyroidism in previously euthyroid children during GH therapy has been reported with widely varying incidence. We monitored the acute effects on the hypothalamic-pituitary-thyroid axis in 15 euthyroid children with classic GH deficiency during the first year of GH therapy. All were initially euthyroid, as assessed by normal baseline TSH, T4, free T4, and T3 levels and negative antithyroid antibodies. A thyroid profile (T4, free T4 index, T3, rT3, and TSH) was performed at baseline and 1, 3, 6, 9, and 12-15 months after GH therapy began; a TRH stimulation test was performed at baseline and after 1, 3, and 9 months of therapy. By 1 month, there were significant decreases in T4, free T4 index, and rT3, and significant increases in T3 and the T3/T4 ratio. The changes from baseline values were greatest at 1 month, were almost universal for all thyroid values, and showed a gradual return to baseline from 3-12 months. There were no clinical signs of hypothyroidism and no change in baseline or TRH-stimulated TSH levels or in cholesterol levels, and all patients grew at velocities expected for the treatment schedule. There is little evidence for the development of clinically significant hypothyroidism in the great majority of initially euthyroid patients after GH therapy is begun. T4 supplementation is seldom needed in such patients.

Influence of genetic and environmental factors on melanocytic naevi: a lesson from Turner's syndrome.

Females with Turner syndrome (TS) are alleged to have increased numbers of melanocytic naevi. Although a high count of acquired melanocytic naevi (AMN) is one of the major risk factors for melanoma, this malignancy has been reported only rarely in patients with TS. The purpose of this study was to explore the effects of environmental and genetic factors on AMN count and density in TS. AMN count and density in 24 patients with TS treated with growth hormone (GH). 24 GH-treated females with GH deficiency (GHD) and 24 normal females were compared in a cross-sectional study. The average AMN density in TS was 50 naevi/m2 as compared with 18 naevi/m2 in the GHD group and 24 naevi/m2 in normal controls (P = 0.001 and P = 0.004, respectively). Duration of GH therapy did not correlate with AMN count (P = 0.44) or AMN density (P = 0.81). The pattern of distribution of naevi between constantly exposed, intermittently exposed and unexposed skin was similar in all groups. Sun exposure was the major factor that affected the regional AMN densities in the control groups, but not in the TS group. The findings of our study indicate that the effects of environmental factors on AMN count and density may vary among genetically different populations. A review of the literature suggested that melanoma is no more prevalent in TS than in the general population.

Lack of effect of growth hormone therapy on the count and density of melanocytic naevi in children.

An observation of accelerated growth of acquired melanocytic naevi (AMN) during treatment with human growth hormone (GH) raised concerns about the potential risk of melanoma in treated patients. An increased number of AMN, rather than growth rate, is associated with a higher risk for melanoma. It is unknown whether treatment with GH causes an increase in numbers of AMN. We evaluated the effect of GH treatment on the number of AMN in a cross-sectional study of 90 children with GH deficiency. AMN counts and densities in these children were compared with those found in a control group of 100 children. Factors potentially related to increased numbers of AMN, such as age, sex, skin colour, number of episodes of sunburn and duration of GH therapy were determined. Among the various factors, only the age and colour of unexposed skin area were predictive for the total number and density of AMN. No correlation was found between the AMN counts or density and the duration of GH therapy. There was no difference in AMN counts or density between the GH-deficient patient group and the control groups. We conclude that GH therapy in children is not associated with increased AMN count and density and is unlikely to potentiate the risk for melanoma in these children.

Widespread growth retardation and variable growth recovery in foster children in the first year after initial placement.

OBJECTIVES: To determine children's growth patterns in the first year of foster care placement and to compare catch-up growth with initial height percentile as indicators of prior growth retardation. DESIGN: Inception cohort. SUBJECTS: Forty-five children aged 1 1/2 to 6.0 years in their first year of foster care. SETTING: Urban, community-based primary care center. MAIN OUTCOME MEASURES: Height, weight, weight-for-height, and annual growth velocity z scores 1 year after placement. RESULTS: The group entered foster care with an overall height deficit (height z = -0.21), grew at an above-average rate (velocity z = +0.33), and eliminated the height deficit by the end of the year (height z = -0.02; P < .05). Weight increased (baseline weight z = -0.16; year-end weight z = +0.35) and correlated with height z change (r = 0.385, P = .009). Weight for expected weight-for-height-age was above average and did not change (baseline weight for expected weight-for-height-age z = +0.30; year-end weight for expected weight-for-height-age z = +0.40). Baseline age correlated with velocity z (r = .413, P = .005) but not with change in height z. Baseline height z did not correlate with either velocity z or change in height z. Three patterns of growth were seen: 21 (47%) showed catch-up growth (height velocity z = 1.34; gain in height z = +0.61); 16 (36%) showed stable growth; and 8 (18%) showed poor growth (height velocity z = -1.49; decrease in height z = -0.49). CONCLUSIONS: Almost half of the children showed significant catch-up growth in the first year after foster care placement, indicating probable prior growth failure. Initial height was not predictive of future growth, and simple screening (such as height less than the fifth percentile) would have missed the majority of children who showed catch-up growth. A substantial minority (18%) continued to decline across height percentiles after placement. The initial and subsequent growth failure and catch-up growth in this population did not appear to be related to nutritional changes.

Survey of growth hormone treatment practices by 251 pediatric endocrinologists.

We wished to determine current GH treatment practices of pediatric endocrinologists and to see whether those practices related to physician characteristics. We analyzed questionnaires completed by 251 of 413 (61%) pediatric endocrinologists attending a national meeting focused on growth research. In general, laboratory testing is little used in deciding to begin or to end GH therapy. Auxological criteria account for 6 of 8 decision items always used by more than 50% of physicians for starting GH treatment and for 5 of 6 items always used for stopping therapy. Although 80% of respondents use 2 GH stimulation tests, only 32% believe such tests predict the response to therapy, 40% do not know which type of assay their lab uses, and 82% use GH in short, poorly growing children regardless of stimulation tests results. Ten percent treat short, normally growing children who pass GH stimulation tests. The median number of syndromes treated off-label was 6. There were no striking differences between faculty and private practitioners in the use of laboratory screening tests, in the use of auxological or laboratory criteria, in perceptions of risk of therapy, or in the number of syndromes treated, nor were there significant differences based on practice volume. Respondents support growth screening in schools and guidelines for the uniform use of GH treatment. Despite the ambiguities and controversies in current GH therapy, pediatric endocrinologists share many diagnostic and therapeutic philosophies.

Clinical and/or biochemical pancreatitis in diabetic ketoacidosis.

The association between diabetic ketoacidosis and acute pancreatitis is examined with reference to four patients. Hypertriglyceridemia was noted in only one of the patients. The pancreatitis was relatively mild and resolved rapidly.

Age-dependent changes in thiamin concentrations in whole blood and cerebrospinal fluid in infants and children.

We determined pediatric age-specific normative ranges for total, phosphorylated, and nonphosphorylated thiamin in whole blood (n = 323) and cerebrospinal fluid (CSF; n = 208). Whole-blood total thiamin decreased from 258 +/- 63 (mean +/- SD; age 0-3 mo group) to 214 +/- 44 nmol/L (age 3-12 mo group) in the first year of life and was stable at 187 +/- 39 nmol/L after 12 mo of age. The overall decline in whole-blood total thiamin was mainly due to a drop in phosphorylated thiamin, the biologically active form. Mean CSF total thiamin decreased from 135 +/- 42 (age 0-9-mo group) to 107 +/- 34 nmol/L (age 9-18-mo group) in the first 1.5 y of life and was stable at 84 +/- 51 nmol/L thereafter. This overall decline was due initially to a drop in nonphosphorylated thiamin and later to a drop in phosphorylated thiamin. The changes in whole blood and CSF occurred independently and probably represent metabolic and neurological maturation of the infant. Whole-blood total and phosphorylated thiamin concentrations were lower in blacks only after pubarche. Age-specific norms should be used for determining the thiamin status in infancy because thiamin concentrations are significantly higher in the first few months of life.

Autoimmune thyroiditis after bone marrow transplantation.

We describe the transfer of autoimmune thyroiditis by bone marrow transplantation. The male recipient developed compensated hypothyroidism 3 years after transplantation, and progressed to thyroid failure 1 year later. The female sibling had compensated hypothyroidism when screened 4 years after marrow donation. The donor, the recipient, and an older male sibling were HLA DR4, 5; the father was presumed homozygous for DR4 and the mother for DR5. Only the donor and recipient had abnormal thyroid function and positive antithyroid antimicrosomal antibodies. The most probable mechanism for the organ specific autoimmune dysfunction is the transfer of abnormal B and T cell clones from the donor. Radiation damage to the thyroid prior to transplantation, the mild graft-versus-host disease immediately following transplantation, and the general immune dysregulation seen in the first several years after marrow engraftment may have contributed to an acceleration of the autoimmune destruction of the recipient's thyroid.

Factors affecting a cyanogen bromide-based assay of thiamin.

We analyzed extensively a modified thiochrome method for thiamin analysis. Acid phosphatase (EC 3.1.3.2) from potato was superior to either alpha-amylase or acid phosphatase from wheat germ as a dephosphorylating agent. Timing of cyanogen bromide exposure was important, but the assay had good precision and accuracy. The standard curve was linear from 10 to 3000 nmol/L. The within-run and between-run coefficients of variation for total thiamin in whole blood were 3.6% and 7.4%, respectively. Analytical recoveries for low, intermediate, and high additions of thiamin to whole blood were 93-109%. Sample yield was increased by 41% (+/- 29% SD) with pre-assay freezing. Samples were stable for two days at room temperature, for seven days when refrigerated, and for two years when frozen. Previously unreported interference was seen with penicillin derivatives, and with several commonly used diuretic and antiepileptic medications. This assay may be suitable for population screening; 200 samples could be analyzed weekly at a cost of +0.20 per sample.

Thiamine status in urban adolescents: effects of race.

To evaluate thiamine status in an urban adolescent population, we performed two investigations. In Study I, we compared whole blood thiamine levels in 101 healthy adolescents from varied racial backgrounds with those that had been obtained previously in 146 healthy white adults from a different geographic locale. Blood thiamine values were significantly lower in the adolescents as a group, but the differences were entirely due to the lower levels in the black adolescents. To explore further these differences (Study II), we compared thiamine status in 34 adolescents with that of their parents using measures of both whole blood thiamine content and of erythrocyte transketolase activity. White adolescents had significantly higher total whole blood thiamine values than black adolescents, and white parents had significantly higher thiamine values than black parents by both total whole blood assay and level of transketolase activity. There were no differences in thiamine status between adolescents and parents of the same race. Racial composition is an important variable to consider in population surveys of thiamine status.

Factors associated with brain herniation in the treatment of diabetic ketoacidosis.

To determine factors contributing to life-threatening brain herniation in patients treated for severe diabetic ketoacidosis, we analyzed history, laboratory data, rate and composition of fluid and insulin administration, and time to onset of brain herniation in nine new cases and 33 prior reports. The overall rate of fluid administration was inversely correlated with the time of onset of herniation (r = -0.32, p = 0.04). Only 4 of 40 cases occurred at fluid intakes less than or equal to 4.0 L/m2/day. During treatment, "calculated" serum sodium concentrations fell significantly and were less than 130 mEq/L in 33% of cases at the time of herniation. These data indicate that excessive secretion of vasopressin may exacerbate the brain edema, and that limitation of the rate of fluid administration may be prudent.