Impact of insulin pumps on glycaemic control in a pump-naïve paediatric regional population


  • Competing interests: The authors have nothing to declare.

Dr Martin de Bock, The Liggins Institute, University of Auckland, Private Bag 92019, Auckland 1023, New Zealand. Fax: +64 9 373 7497; email:


Aim:  To examine the clinical impact of insulin-pump therapy for children with type 1 diabetes mellitus (T1DM) in a regional paediatric service, Auckland, New Zealand.

Methods:  Retrospective analysis of children with T1DM from the Starship paediatric diabetes database who started on insulin-pump therapy from 2002 to 2008 compared with the whole T1DM population and with an equal number of non-pump patients matched by age, sex, ethnicity and duration of diabetes.

Results:  From 621 subjects with 6680 clinic visits, 75 children were treated with insulin-pump therapy for more than 12 months. Transitioning to insulin-pump treatment was associated with an improvement in HbA1c compared with baseline (−0.3%/year, P < 0.001) for up to 3 years. In contrast, despite similar deprivation scores, non-pump controls showed a continuing trend to higher HbA1C values (+0.2%/year, P < 0.01). The risk of severe hypoglycaemia fell after pump start (from 27 (0–223) to 5 (0–0.91) events/100 patient years) with no change in non-pump controls; the rate of diabetic ketoacidosis remained low in both groups.

Conclusions:  In a pump-naïve regional paediatric population, insulin-pump therapy for T1DM was safe and effective, and associated with sustained improvements in HbA1c and lower risk of hypoglycaemia.