Article Text

Neonatal hyperinsulinism

    Statistics from

    Neonatal hyperinsulinism is caused by either diffuse β cell hyperplasia (nesidioblastosis) or focal adenomatous islet cell hyperplasia. The former condition has several known causes such as mutation of the sulphonylurea receptor gene or of the potassium channel gene (KIR 6.2), both recessively inherited, or of the glucokinase and other genes (dominant). A mutation of the glutamate dehydrogenase gene causes both hyperinsulinism and hyperammonaemia. Focal adenomatous islet cell hyperplasia is an example of genetic imprinting and is associated with loss of the maternal copy of the sulphonylurea receptor type 1 gene on chromosome 11 (11p 15). Workers at the Hôpital des Enfants Malades in Paris (Pascale de Lonlay-Debeney and colleagues. New England Journal of Medicine 1999;340:1169–75) have described their management of 52 neonates and their ability to distinguish between the two types of hyperinsulinism and, as a consequence, to perform restricted surgery when appropriate.

     Thirty babies proved at operation to have diffuse hyperplasia and 22 to have focal adenomatous hyperplasia. The clinical features in the two groups were similar, glucose infusion rates needed to maintain plasma glucose of 3.0 mmol/l or more were around 16 mg/kg/min, and surgery was performed at around 3 months. Diazoxide therapy produced little benefit.

     Preoperative assessment included transhepatic catheterisation of the portal vein and selective catheterisation of the pancreatic vein with removal of blood from selective sites within the pancreas for measurement of plasma glucose, insulin, and C peptide. This was done in 26 of the 30 in the diffuse group and 19 of the 22 in the focal group. As a result, in the diffuse hyperplasia group, 17 infants were shown to have insulin hypersecretion from all areas of the pancreas, localised hypersecretion was suspected in seven, and the results were inconclusive in two. In the focal adenomatous hyperplasia group, localised hypersecretion was detected in 17 of the 19 tested, and the localisation in each case was confirmed by immediate histology of biopsy specimens at operation. The two infants in this group with inconclusive preoperative studies both had lesions confined to the body of the pancreas. The location of the lesions in the 22 infants was head (n = 9), isthmus (n = 3), body (n = 8), and tail (n = 2).

     All 30 infants with diffuse hyperplasia underwent near total pancreatectomy, whereas this operation was performed on only three in the focal adenomatous hyperplasia group (the three who did not have preoperative catheter studies), the rest had partial pancreatectomy. After near total pancreatectomy for diffuse disease, 13 children had persisting hypoglycaemia, eight developed insulin dependent diabetes, and seven had abnormal glucose tolerance. After partial pancreatectomy for focal disease there was no hypoglycaemia and glucose metabolism was normal.

     Preoperative studies of localised pancreatic insulin secretion together with intraoperative histology can allow identification of patients suitable for partial pancreatectomy with much improved results.

    View Abstract

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.