49 centres in Austria, France, Germany, and Italy. Eligible patients had cystic fibrosis, were older than 10 years, and had newly diagnosed diabetes.
Interventions
Insulin or repaglinide, stratified by sex and age (10-15 years or >15 years). 34 patients in the repaglinide group and 41 in the insulin group
Outcome measures
The primary outcome was glycaemic control assessed by mean change in HbA1c concentration from baseline after 24 months of treatment. Differences between groups were assessed by linear models.
Main results
At 24 months, glycaemic control was similar in the repaglinide and insulin groups (mean change in HbA1c concentration from baseline 0.2% [SD 0.7%], 1.7 mmol/mol [8.1 mmol/mol] with repaglinide vs -0.2% [1.3%], -2.7 mmol/mol, [14.5 mmol/mol] with insulin; mean difference between groups -0.4%, (95% CI -1.1 to 0.2 [-4.4 mmol/mol, -11.5 to 2.7], p=0.15). The most frequent adverse events were pulmonary events (43 [40%] of 107 in the repaglinide group and 60 [45%] of 133 in the insulin group), and the most frequent serious adverse events were pulmonary events leading to hospital admission (five [50%] of ten and seven [54%] of 13, respectively).
Authors' conclusions
Repaglinide for glycaemic control in patients with cystic-fibrosis-related diabetes is as efficacious and safe as insulin.