The aim of our trial is to try to switch patients with permanent neonatal diabetes mellitus due to a Kir6.2 or SUR1 activating mutation from subcutaneous insulin to oral glibenclamide therapy.

Neonatal diabetes mellitus, characterized by hyperglycaemia requiring exogenous insulin therapy appearing during the first months of life, is a rare condition with an estimated incidence of 1 in 400000 newborns and is permanent in only one-half of the patients[1]. Several studies have identified heterozygous activating mutations of the coding sequence of KCNJ11 or ABCC8 in patients having a permanent neonatal diabetes mellitus [5,6,7,8]. These genes encode for the 2 type of subunits Kir6.2 or SUR1 of the pancreatic β-cell ATP-sensitive K+ channel (KATP channel) which plays a central role in glucose-stimulating insulin secretion. These channels are also found on muscle and nervous cells, and this may explain the neurological features sometimes associated with permanent neonatal diabetes mellitus. Some sulfonylureas, as the glibenclamide, stimulate insulin secretion by binding to SUR1 subunit and closing KATP channels by an ATP-independent mechanism. The glibenclamide is used efficiently in type 2 diabetes but also recently in replacement of subcutaneous injected insulin in children with a Kir6.2 or SUR1 activating mutation [7,8,11-13].

The aim of our trial is to try to switch patients with permanent neonatal diabetes mellitus due to a Kir6.2 or SUR1 activating mutation from subcutaneous insulin to oral glibenclamide therapy. This study will stand at Necker-Enfants Malades Hospital in the Endocrinology and Diabetology Unit of the Professors Robert and POLAK. It will include 20 patients, most of them already identified. This study has two purposes: therapeutic by switching the patients from subcutaneous insulin to oral glibenclamide therapy, and cognitive by a complementary evaluation and understanding of the mechanisms of insulin secretion and of glibenclamide efficiency. To do so, we will assess continuously the capillary glycaemia for three consecutive days and evaluate the insulin secretion under insulin and sulfonylureas. Furthermore, we will rate the neurological and developmental status of the patients to seek for a potential improvement under glibenclamide therapy.

If oral glibenclamide therapy for these patients is proved to be successful, the systematic search for a Kir6.2 or SUR1 activating heterozygous mutation in newborns with permanent neonatal diabetes mellitus could be recommended in order to start early oral glibenclamide therapy and thus extend the indications for the sulfonylureas.

Inclusion Criteria:

• coding sequence of KCNJ11 or ABCC8 in patients having a permanent neonatal diabetes mellitus
• written informed consent

Exclusion Criteria:

• hypersensibility of sulfonylureas
• severe renal failure (clearance of creatinemia < 30 ml/min)
• severe hepatic failure (Prothrombin rate < 70 %)
• Porphyria
• imidazol treatments
• pregnancy
• no social security affiliation