Assessing effect of mixing insulins by glucose-clamp technique in subjects with diabetes mellitus.

The glucose-clamp technique was used to assess the effect on insulin activity of mixing clinically relevant doses of short- and longer-acting insulins in insulin-treated diabetic subjects. Mixtures of porcine regular and lente insulins resulted in a reduction in activity of the mixture compared with separate injection that was more apparent with premixing for 5 min before injecting than with premixing for 2 min. These findings were not explained by differences in volume of the injected insulin preparations. Mixing of porcine regular and bovine ultralente insulins for 2 min before injecting resulted in a reduced activity compared with separate injection. No difference in activity was obvious for porcine regular and NPH insulins given separately or premixed for 5 min. A reduction of activity during the 1st h after injection was observed when separate injection of one brand of these insulins (Velosulin or Insulatard) was compared with the manufactured mixture (Mixtard). These results indicate that premixing regular and NPH insulins in a 1-to-2 ratio does not alter the biological activity compared with separate administration of the insulins. However, the mixing of regular and insulin-zinc suspensions results in a loss of insulin activity, the magnitude of which depends on the time between mixing and injecting the insulin. The clinical significance of the effect of mixing on the efficacy of subcutaneous insulin therapy should be considered in the context that this is only one of many factors that may affect the activity of subcutaneously injected insulin.