D0 AUTOANTIBODIES INFLUENCE THE CLINICAL COURSE AND OUTCOME OF TYPE I DIABETES?
  S Hofer 1, E Schober 2 A Thon 3, O Seewi 4, G Däublin 5, U Krause 6, RW Holl 6 on behalf of the DPV-Wiss-Studygroup
1 Medical Universtity of Innsbruck, Department of Paediatrics, 2 Medical University of Vienna, Department of Paediatrics, 3 University of Hannover, Children´s Hospital, 4 University of Köln, Children´s Hospital, 5 Children´s hospital in Aurich , 6 Universtiy of Ulm, Department of Epidemiology.
  The aim of the study: Patients with type I diabetes are divided into antibody positive – autoimmune diabetes and antibody negative patients – idiopathic diabetes.

The aim of this multicenter retrospective study is to investigate the influence of antibodies on the course and the long term outcome of patients with type I diabetes.

Methods: The DPV-Wiss-Project consists of anonymous data of patients with diabetes. Over a time period of 13 years (1990 and 2003) 7017 paediatric patients with diabetes were recorded in 150 centres in Austria and Germany. Paediatric patients were defined as younger than 20 years of age. The antibodies measured were insulin antibodies, islet cell antibodies and glutamic acid decarboxylase antibodies.

Patients with at least one antibody positive at diagnosis were defined as antibody-positive and type 1A diabetes, patients with no antibodies detectable at diagnosis were defined as antibody-negative and type 1B diabetes. Patients with missing documentation were excluded from the analysis.

Results: 5206 (74.2%) patients were classified as type 1A diabetes, 1811 (25.8%) as type 1B diabetes.

Patients with detectable autoimmunity were older at the time of diagnosis and significantly more often female. Within the first six years of diabetes history patients classified as type 1A needed more insulin per kg bodyweight. After adjusting the data for age, sex and diabetes duration no significant differences between type 1A and type 1B diabetes could be identified.

We therefore conclude that the presence of antibodies at diagnosis is not a predictive factor for the course of type 1 diabetes in children and adolescents.