Screening for coeliac disease (CD) with serum antigliadin antibodies (AGA) was performed in 1032 diabetic children and adolescents. In 8 children CD had been diagnosed before study entry. Of the remaining 1024 children, 33 had an elevated AGA titre in the first serum sample. On follow-up an elevated AGA titre was confirmed in only 17 of 31 patients. Nine of the repeatedly positive patients underwent jejunal biopsy, and CD was diagnosed in two asymptomatic patients; both were positive for IgG- and IgA-AGA. Among 10 AGA-positive patients in whom biopsies could not be performed, only 1 showed IgA-AGA and thus carried a high risk for CD. From our results we estimate a prevalence of CD in Swiss and German diabetic children between 1.1% and 1.3%. False-positive AGA titres occurred significantly more often in patients with diabetes duration of less than 1 year. AGA testing reached a specificity of 99% if performed at least 1 year after the onset of diabetes. Children suffering from both diabetes and CD showed a diabetes manifestation at a significantly younger age than non-coeliac patients, whereas CD tended to be diagnosed at a remarkably late age.

The presence of HLA-DR 3 was analysed in 745 patients with Type 1 (insulin-dependent) diabetes with age at diagnosis between 1-19 years. HLA-DR 3 and/or 4 was found in 678/745 (91%) of the patients. Presence of DR2 with neither DR 3 nor 4 was demonstrated in 15 patients. Patients with HLA-DR 3 without DR 4 presented with Type 1 diabetes more evenly over the year; they also presented without incidence peaks at 7 years or 10-11 years, as seen especially in DR 3/4 patients. The DR 3 patients more often had mild disease with less ketonuria at diagnosis, less often ketoacidotic symptoms and more often a subsequent partial remission. The apparently more severe disease among diabetic girls may, at least to some extent, be explained by their higher prevalence of HLA-DR 4. The differences found were similar in North America and Europe. The results suggest that Type 1 diabetes is a genetically heterogeneous disease and that HLA-typing may be a useful marker of this heterogeneity.

No abstract

A multicenter, longitudinal study of children below the age of 16 years with newly diagnosed Type 1 (insulin-dependent) diabetes treated either with porcine monocomponent insulin (n = 26) or semisynthetic human monocomponent insulin (n = 26) was performed during the first 24 months after onset of diabetes. The two groups were carefully matched for age, duration of disease symptoms, initial metabolic values, islet cell antibodies and HLA-DR antigens. During the 24-month observation period there was no significant difference between the two groups in respect to the clinical course, insulin dosage, HbA1 and residual B-cell activity. No child in either group had a real remission without necessitating insulin therapy. The prevalence of insulin antibodies increased slowly and was 62% in the group treated by human insulin and 52% in the porcine insulin-treated group after 24 months. The titres were generally low and there was no statistical difference between the two groups in respect to insulin antibody formation.

224Short Communications 96% of the radioactivity in A could be precipitated by incubation with an excess of rabbit anti-glucagon serum and subsequent addition of ethanoCln a corresponding test without antiserum, 3-4% of the radioactivity was precipitated.The relative amounts of 125 1 in r, MIT and DIT were found to be 2.6, 97.2 and 0.3%, respectively. The corresponding vaiues for a material radioiodinated in the same way as the material applied on the column but only purified by a saltingout procedure, were found to be 1.8,83.9 and 14.3%, respectively (the small amount of 125r found in both cases were probably due to damage sustained during storage). Thus, one effect of the anion exchange chromatography and subsequent selection of the ~-pool was a practically complete rem oval of DIT-containing material.It is concluded that essentially all of the 125 1 in ~ was present as MIT-iodine in immunoreactive 1251-glucagon. We have not investigated how the radioiodine was distributed between the two tyrosines of the glucagon molecule. Different dilutions of A and of another 1251-glucagon (B), prepared by the same iodination procedure but followed only bya crude purification-as described by Heding (1971), were used for making standard curves in the glucagon radioimmunoassay. The two sets of curves did not cross each other at any point. Fig. 1 b shows optimal standard curves for ~ and !! with the same rabbit anti-glucagon serum. It appears that the 1251-glucagon purified by chromatography was better suited for use in the radioimmunoassay than the crudely purified 125I-glucagon. This finding has been reproduced with anti-glucagon sera from different rabbits and with different preparations of each of the two kinds of 12SI-glucagon.The nucleotide metabolism of the organism and the effect of the nucleotides on various metabolie processes are not yet clear. In view of the similarity between the biguanide derivates used with success in some cases of diabetes mellitus and some nucleotides, this problem is again, beginning to arouse interest, and the possibility of naturally occurring guanido compounds in the regulation of carbohydrate and protein metabolism has been propounded (Butterjield, Cox and Whichelow 1968).In a previous paper we demonstrated that adenine and guarIine increase the glucose release of rat Iiver perfused with glucose-Downloaded by: National University of Singapore. Copyrighted material.