The clinical effects of mixing short- and intermediate-acting insulins in the treatment of non-insulin-dependent diabetes

Premixing short- and intermediate-acting insulins in one syringe, with refrigerated storage before injection, is practised by some centres in the treatment of older patients with non-insulin-dependent diabetes. Because other studies have reported the loss of the short-acting insulin component after mixing with intermediate-acting insulins, we examined the clinical effect of mixing soluble insulin with lente or isophane insulins in subjects with non-insulin-dependent diabetes. When soluble and lente insulins were mixed in the same syringe and injected immediately, the peak level of insulin was very similar to the peak level after separate injections but occurred at five hours instead of three hours after the injection. As a result, the plasma free-insulin profile over three hours was lower with premixed insulin than after separate injections of the two insulins (incremental insulin area, 88 ± 20 mU·L^-1·h, and 129 ± 37 mU·L^-1·h, respectively; P < 0.05). This delay in the absorption of soluble insulin caused a greater rise in plasma glucose levels such that the incremental glucose area over eight hours was 25.5 ± 4.4 mmol·L^-1·h for premixed insulin compared with 10.4 ± 6.2 mmol·L^-1·h for separate injections (P < 0.05). Soluble and isophane insulins had similar absorption profiles whether injected separately or premixed (incremental insulin area, 0 to 3 h, 176 ± 44 mU·L^-1·h and 156 ± 29 mU·L^-1·h, respectively). Our results indicate that the absorption of soluble insulin is delayed when it is mixed with lente insulin but not with isophane insulin. Even in subjects with endogenous insulin secretion, this effect may have clinical importance and should be taken into account when insulin therapy is adjusted for patients with non-insulin-dependent diabetes.