The chlorpropamide alcohol flush. Lack of specificity for familial non-insulin dependent diabetes.

A facial flush provoked by alcohol in chlorpropamide treated diabetics has been described as a genetic marker for a dominantly inherited type of non-insulin dependent diabetes. In this study a chlorpropamide alcohol flush was observed in 16.9% of control subjects (n = 154), 23.3% of insulin dependent diabetics (n = 437) and 16.5% of patients with non-insulin dependent diabetes (n = 145). Among the non-insulin dependent diabetics no difference in the frequency of the chlorpropamide alcohol flush was found between those with and without a family history of diabetes. Specificity was not improved by skin temperature measurement or additional placebo tests. According to these data the chlorpropamide alcohol flush does not seem to be specific for non-insulin dependent diabetes and hypotheses about the aetiology of this type of diabetes based on the chlorpropamide alcohol flush should be regarded with caution.

HLA-typing in juvenile diabetics with and without positive family history and in families with one and two diabetic siblings.

HLA-typing was performed in two groups of juvenile-onset diabetics, one with (n = 58) and one without (n = 109) a family history of the disease. The association of this type of diabetes with certain HLA antigens (excess of B8 and B15, shortage of B7) was confirmed. No heterogeneities could be established between the two groups. This suggests that the aetiologic basis in single and familial cases of juvenile diabetes is the same. The hypothesis, that the B8 associated gene is more penetrant than the B15 associated gene, cannot be confirmed. Haplotypes were determined in families with one and two diabetic siblings. The findings of high haplotype concordance among diabetic siblings was confirmed: concordance of 2, 1 and 0 haplotypes in 7, 5 and 3 pairs respectively. There was a low degree of haplotype concordance between diabetics and nonaffected siblings in the families with two diabetics: 2, 1, and 0 haplotypes in 2, 8 and 6 pairs respectively. This led to the hypothesis of negative selection against these HLA-linked "diabetogenic" genes. This tendency was not, however, observed in families with only one diabetic. The report of a high recombination rate in families with juvenile diabetics could not be confirmed.