Normal proinsulin processing despite beta-cell dysfunction in persistent hyperinsulinaemic hypoglycaemia of infancy (nesidioblastosis)

Abstract

Persistent hyperinsulinaemic hypoglycaemia of infancy (PHHI) is a genetic disorder which causes severe hypoglycaemia in the neonate. The beta cells fail to respond to changes in blood glucose levels in all the stages of the disease, which often ends with NIDDM. Fasting insulin, intact proinsulin and des 31,32 split proinsulin levels were measured in PHHI patients with active disease, patients after partial pancreatectomy, and those in clinical remission. All but one of the pancreatectomized patients developed diabetes and were hyperglycaemic on evaluation. Fasting insulin was comparable in pancreatectomized and medically treated patients. Des 31,32 split proinsulin levels were much higher in pancre atectomized compared to non-pancreatectomized patients (10.7 ± 2.5 vs 3.4 ± 0.8 pmol/l, p = 0.001) and age-matched control subjects (3.8 ± 1.4 pmol/l, p = 0.018). Also the ratio of des 31,32 split proinsulin to total insulin plus proinsulin-like peptides was higher in pancreatectomized patients (18.7 ± 2.8 vs 7.2 ± 0.8% in non-pancreatectomized patients, p = 0.001, and 6.8 ± 2.1% in normal control subjects, p = 0.004). Furthermore, des 31,32 split proinsulin was the dominating species of proinsulin-like molecules in the pancreatectomized patients (62.7 ± 1.6% vs 45.5 ± 3.5%, and 49.0 ± 3.2% in non-pancreatectomized patients and control subjects, respectively, p = 0.001 and p = 0.0002). Intact proinsulin levels, and the proinsulin percentage, tended to be higher in pancreatectomized patients; however, the differences did not reach statistical significance. All parameters were similar in non-pancreatectomized patients and age-matched control subjects. Subgroup analysis showed comparable proinsulin-like peptide levels in patients with active disease and those in apparent clinical remission. Fasting levels of insulin and proinsulin-like peptides were also measured in a larger group of healthy children and young adults. Insulin and des 31,32 split proinsulin increased with age, the differences being most prominent when the young age group (0-8 years) was compared to the older groups (8-16 and > 16 years). The fasting levels of plasma insulin were correlated with those of intact proinsulin and des 31,32 split proinsulin (r = 0.82 and 0.81, respectively). Fasting insulin, intact proinsulin and des 31,32 split proinsulin were correlated with BMI (r = 0.55, 0.56 and 0.53, respectively). In summary, relative hyperproinsulinaemia was noted only in PHHI patients with increased secretory demand following pancreatectomy, but not in patients with active disease or those in spontaneous clinical remission. These findings suggest that abnormal proinsulin processing is not an intrinsic feature of PHHI despite the severe beta-cell dysfunction.