Determination of insulin antibodies.

Insulin antibodies were determined as percentage binding of 125I-insulin in the sera of normal persons and of diabetic subjects treated and untreated with insulin. The effect of the dilution of the serum, circulating insulin and extraction of free and total insulin was evaluated. The determination of insulin antibodies in samples at a final dilution of 1:10 clearly discriminated between insulin-treated and untreated subjects. In insulin-treated subjects, the determination of insulin antibodies in samples at a final dilution of 1:100 gave false-negative results in 28 per cent. However, the determination of insulin antibodies at a final dilution of 1:100 discriminated between insulin-resistant and non-resistant diabetic subjects. Extraction of total insulin at pH 3.0 using 0.1 N HCl increased the percentage of 125I-insulin binding significantly. Extraction of free insulin by charcoal from the samples did not increase the binding of 125I-insulin. The injection of crystalline insulin 4 hours prior to withdrawing the samples did not decrease binding of 125I-insulin.