Objective

To determine whether there are different rates of partial remission in preschool, school-age children, and adolescents with type 1 diabetes mellitus (T1DM) and to identify clinical characteristics that are associated with increased rate of partial remission.

Design/methods

A total of 152 consecutive patients with newly diagnosed T1DM in 2004 were studied. Clinical characteristics at diagnosis, hemoglobin A1C (HbA1C), and total daily insulin dose (TDD) at 3-month interval follow-up for 1 yr were analyzed in each age-group (group 1, aged <5 yr; group 2, aged 5-12 yr; and group 3, aged >12 yr). Partial remission was defined as TDD <0.5 units/kg/d with HbA1C <8% assessed at 6 months after diagnosis.

Results

Young children (group 1, 26.8%) and adolescents (group 3, 29%) had low rates of partial remission compared with school-age children (group 2, 56%, p = 0.002). There were no differences in the rates of diabetic ketoacidosis (DKA), autoantibody frequency, and HbA1C at diagnosis between age-groups. DKA at diagnosis was associated with less likelihood of having partial remission (p < 0.001). There were no associations between gender, autoantibodies, and HbA1C at diagnosis and the rate of partial remission.

Conclusions

Young children and adolescent children with T1DM had a low rate of partial remission. Metabolic control was poorest in young children, whereas higher dose insulin in adolescents because of insulin resistance contributes to less likelihood of having partial remission. DKA at diagnosis was associated with low rate of partial remission. It is possible that the low frequency of honeymoon phase in young children reflects more aggressive beta-cell destruction in young children.