Insulin-pump use in everyday practice: data from an exhaustive regional registry in France.

Aim: The aim of this study is to evaluate the effectiveness and safety of continuous subcutaneous insulin infusion (CSII) under real-life conditions among all patients treated with CSII in the south of Paris.

Methods: The 42 diabetologists practising in the region enrolled all patients treated with CSII or admitted for CSII initiation. During the study visit, the data for pump use and clinical results were recorded.

Results: Data were obtained for 424 patients, mean age 44.2+/-15.6 years, disease duration 18.7+/-10.6 years, including 339 treated with CSII for longer than three months (mean duration: 3.5+/-3.5 years; range: 3-258 months). Most of the patients (N=285, 84.8%) had type 1 diabetes; 44 (13.1%) had type 2 diabetes. In patients treated for more than three months, HbA1c decreased significantly between CSII initiation (9.1+/-1.9%) and the study visit (7.8+/-1.4%; P<0.0001). Patients with HbA1c >9%, using the pump, experienced a significant 0.9% improvement in their HbA1c levels with CSII versus multiple daily injections (P=0.001). The number of episodes of moderate hypoglycaemia was 2.7+/-2.5 per patient per week; of severe hypoglycaemia, 0.34 per patient per year and of ketoacidosis, 0.11 per patient per year. Factors significantly associated with HbA1c levels included amount of physical activity, pregnancy, HbA1c at CSII initiation and number of glucose self-determinations. Those associated with the number of moderate hypoglycaemia episodes were basal rate number, female gender and HbA1c level. HbA1c was negatively correlated with moderate hypoglycaemia (P<0.001), but not with severe hypoglycaemia.

Conclusion: This 'pump' registry establishes the effectiveness of CSII in everyday practice, yet underscores the risks of severe hypoglycaemia and ketosis episodes. It could help diabetologists to improve patient training programmes and follow-up.