Hypersecretion of proinsulin does not explain the hyperinsulinaemia of patients with liver cirrhosis.

A radioimmunoassay using a proinsulin-specific antiserum that does not react preferentially with the split forms of proinsulin has been used to compare the response of circulating proinsulin to low (25 g) and high (75 g) oral glucose loads in healthy subjects and in patients with liver cirrhosis. The patients were divided into two groups: Group A (n = 7) with normal glucose tolerance and Group B with diabetic (n = 5) and impaired (n = 1) glucose tolerance. There was no apparent correlation between glucose tolerance and the results of quantitative liver function tests. In the fasted state, the concentrations of serum proinsulin did not differ significantly in patients of Group A (0.022 +/- 0.002 nmol/l) or Group B (0.026 +/- 0.004 nmol/l) from those in healthy subjects (0.021 +/- 0.002 nmol/l). After 75 g glucose, the rise in serum proinsulin to a maximum concentration of 0.082 +/- 0.012 nmol/l in patients of Group A and to 0.070 +/- 0.019 nmol/l in Group B was not significantly different at any time point up to 180 min from the rise in healthy subjects (to 0.063 +/- 0.005 nmol/l). After 25 g glucose, the response of serum proinsulin in Group B patients (maximum concentration 0.035 +/- 0.003 nmol/l) was not significantly different from that in healthy subjects (maximum concentration 0.032 +/- 0.003 nmol/l) but a slightly enhanced release was observed in the Group A patients (maximum concentration 0.049 +/- 0.003 nmol/l) that was significantly greater (P less than 0.05) at 60 min post-glucose. In contrast, the concentrations of serum immunoreactive insulin and immunoreactive C-peptide in all patients with cirrhosis were significantly elevated compared with healthy subjects both in the fasted state and at several time points following high and low oral glucose. In the fasted state, the serum proinsulin/C-peptide molar ratio, an index of the relative state of secretion of proinsulin and insulin, was significantly lower (P less than 0.05) in both groups of cirrhotic patients than in healthy subjects. After high and low glucose, this ratio fell in all patients and in the healthy subjects. We conclude that cirrhosis of the liver is associated with a hypersecretion of insulin but hyperproinsulinaemia does not contribute appreciably to the elevated concentration of immunoreactive insulin in the peripheral circulation.