[C-peptide determination in diabetics for the evaluation of insulin requirements].

45 non-insulin dependent diabetic subjects with "secondary failure" of oral treatment were observed for 2 years; during this period 23 of the 45 became insulin requiring. Plasma C-peptide concentrations at study entry and several clinical parameters of the patients were analysed statistically using discriminant analysis to test their value in predicting the need for later insulin therapy. The results demonstrated the fasting C-peptide concentrations and, independently of this, the simultaneously measured glucose level had a significant predictive power, e.g. only patients with fasting glycemia of greater than 12 mmol/l and C-peptide concentrations exceeding 700 pmol/l did not require insulin during follow-up. Increased C-peptide concentrations pointed to a predominance of insulin resistance which correlated with an increased chance of being treatable without insulin. A further group of 44 elderly, insulin-treated diabetic subjects were examined for clinical markers of type 1 and type 2 diabetes when these patients were separated into type 1 and 2 diabetes according to their fasting C-peptide concentrations. It was demonstrated that age, weight and duration of diabetes did not significantly differ between the 2 types of diabetes in this group of patients. The studies demonstrate that determination of plasma C-peptide concentration is a valuable tool in distinguishing the two types of diabetes (type 1 and type 2). This distinction has clinical implications with regard to genetic counselling or the search for other endocrinopathies. In addition, fasting C-peptide measurement is a valuable tool, in conjunction with blood sugar determination, in deciding whether patients with secondary failure of oral antidiabetic treatment require long-term insulin therapy.