A comparison of the rapid-acting insulin analogue glulisine with lispro and aspart for the pump treatment of patients with type 1 diabetes

Aims (1) To describe the population of patients with type 1 diabetes (T1DM) using the rapid-acting insulin analogue glulisine versus lispro and aspart during continuous subcutaneous insulin infusion (CSII); (2) to describe insulin relative effectiveness based on hemoglobin A1c (HbA1c), fasting blood glucose (FBG) and dose; (3) to determine rates of hyperglycemia, hypoglycemia, and diabetic ketoacidosis (DKA). Methods The analysis used March 2021 data from the Diabetes-Patienten-Verlaufsdokumentation registry, which contains data of 618,903 patients with diabetes. Patients were propensity-matched by age, sex, and diabetes duration. Results Overall, 42,736 patients of any age were eligible for analysis based on insulin pump usage with either glulisine (N = 707) or lispro/aspart (N = 42,029) between 2004 and 2020. Patients receiving glulisine were older (median 20.0 vs. 16.2 years), equally often male (47.2% vs. 47.8%) and had a longer diabetes duration (median 9.4 vs. 7.4 years). After propensity score matching, 707 pairs remained (total N = 1414). Patient characteristics between groups were similar. Achieved HbA1c values were also comparable: 8.04%, 64 mmol/mol versus 7.96%, 63 mmol/mol for glulisine and lispro/aspart [LS mean difference 0.08 (95%CI - 0.08, 0.25)]. FBG was 9.37 mmol/L (168.9 mg/dL) and 9.58 mmol/L (172.6 mg/dL) in the glulisine and lispro/aspart groups [LS mean diff. - 0.21; (95%CI - 1.13, 0.72)]. Total daily insulin doses and prandial to total insulin ratios were also similar. Glulisine group patients had higher rates of lipodystrophy (0.85% vs. 0.71%) (LS mean diff. 0.18 [95% CI - 1.01, 1.38]) and non-severe DKA (3.11% vs. 0.57%; p = 0.002). Fewer patients in the glulisine group had severe hypoglycemic events (7.66 vs. 9.09; p = 0.333) and severe ketoacidosis events (0.57% vs. 1.56%; p = 0.082) but more had hypoglycemic coma events (p = 0.773), although the differences were not statistically significant. Conclusions Insulin glulisine had comparable glucose control to lispro/aspart. The use of glulisine was less frequent in the present analysis compared to the previous trials.