Insulin autoimmune syndrome in a child with type 1 diabetes: Clinical, Immunological, and Biochemical correlations

Abstract

Introduction: Insulin Autoimmune Syndrome is a rare and incompletely understood cause of hypoglycemia which is characterised by alteration of endogenous or exogenous insulin action by anti-insulin antibodies (IAA). Our patient with Type 1 Diabetes (T1D), like previous case reports, developed life-threatening hypoglycemia. His glycemic control was characterised by delayed insulin action following subcutaneous delivery with resulting progressive hyperglycemia, and then profound hypoglycemia requiring extraordinary amounts of dextrose for rescue therapy. Insulin quantification was affected by IAA binding and, was therefore, performed pre- and post-polyethylene glycol (PEG) precipitation. Treatment with plasmapheresis was aimed at removing immune complexes. Aims: To compare %FII, and glucose level vs %FII in (1) patient, T1D control, and non-T1D control samples and (2) patient samples pre- and post-plasmapheresis. Methods: Plasma glucose and insulin levels were measured using Beckman Coulter assays. Statistical comparisons were performed via Mann-Whitney Test, ANOVA and ANCOVA. Results: The male patient was 9.36 years old with an estimated T1D duration of 7.00 years and HbA1c 7.8% at the time of investigation. He had been managed on various insulin preparations and regimens including multiple daily injections and continuous subcutaneous insulin (CSII). The characteristics of T1D controls (n=12) included: 50% male, 33.3% on CSII, average age of 12.20 ± 3.69 years, average T1D duration of 3.91 ± 0.87 years, and average HbA1c 8.5 ± 0.6%. The non-T1D controls (n=8) were 62.5% male and had an average age of 13.61 ± 0.71 years. The %FII in pre-plasmapheresis (30.3 ± 3.2%) patient samples was significantly lower than T1D control (44.9 ± 7.9%, p<0.05), non-T1D control (106.5 ± 1.6%, p<0.0001), and post-plasmapheresis patient (89.4 ± 1.8%, p<0.0001) samples. The regression slope of glucose vs %FII in patient samples was significantly different from T1D control (p<0.01) and post-plasmapheresis patient (p<0.001) samples. There was no significance difference between the slopes of post-plasmapheresis patient and T1D control samples (p=0.4140). Discussion: Our patient demonstrated lower %FII than T1D and non-T1D controls. Plasmapheresis increased %FII and re-established the glucose to %FII relationship to those observed in T1D controls. This was associated with improved glycaemic predictability and reduced hypoglycaemia frequency.L6135232512016-12-08 <br />