Glycodelin responses to hyperinsulinaemic clamp vary according to basal serum glycodelin concentration

OBJECTIVE: Treatment with metformin, an insulin-lowering agent, increases serum glycodelin, a progesterone-regulated lipocalin protein of the reproductive axis that may play a role in foeto-maternal defence mechanisms. This finding led to the hypothesis that insulin might decrease serum glycodelin concentration. DESIGN, PATIENTS AND MEASUREMENTS: Euglycaemic hyperinsulinaemic clamp experiments (n = 50) were carried out on 28 women of reproductive age (range 25-47 years; mean +/- SEM 39 +/- 1.0 years), and the results were analysed with respect to their baseline serum progesterone (< 10 or > or = 10 nmol/l) and glycodelin (< 10 or > or = 10 microg/l, equivalent to < 357 or > or = 357 pmol/l) concentrations at the onset of the clamp. Ten clamp experiments were performed on five women wearing a levonorgestrel-releasing intrauterine device (IUD), and these were analysed as a separate group. RESULTS: Contrary to the hypothesis, no acute glycodelin-lowering effect of insulin was found in any of the groups studied. All the small rises in glycodelin levels detected during acute hyperinsulinaemia occurred in the comparisons of medians and not means, and all such changes took place within the limits seen in the women with no progesterone exposure. In the group with low progesterone/low glycodelin (n = 21), glycodelin showed a small but significant increase at 30 and 90 min of the clamp (P < 0.01). In the group with elevated progesterone/low glycodelin (n = 11), there was a slight glycodelin increase at 30 min (P < 0.05), whereas no increase was found in the group with elevated glycodelin levels (n = 8). In the clamp experiments on women with levonorgestrel-releasing IUD, the basal glycodelin level was low in all cases and, as in the other women with low glycodelin levels, glycodelin was slightly increased at 30, 60 and 90 min of hyperinsulinaemia (P < 0.01). CONCLUSIONS: The results rule out any acute glycodelin-reducing effects of insulin, although indirect long-term effects mediated by insulin on glycodelin secretion cannot be excluded.