Choosing an insulin requires consideration of species (human, cow or pig), pharmacodynamics (rate of onset and offset and timing of peak effects), administration (dose, frequency, and ability to mix with another formulation), and method of injection (disposable plastic syringe, disposable pen, pump, etc). Pork and beef insulins are highly purified product derived from, respectively, the porcine and bovine pancreas. Human insulin is prepared either by enzymatic modification of porcine insulin, or by recombinant DNA techniques using bacteria or yeast. Human insulin is the hormone of choice because it is thought to be less immunogenic.
Insulin formulations can be short-, intermediate-, or long- acting. Insulin in an unmodified (soluble) form is short- acting with a 30-60 minute onset of action, a peak effect between 1-3 hours, and a duration of up to 8 hours. Lilly’s new insulin lispro (Humalog) has an even shorter onset: 15-30 minutes. Intermediate-acting insulin formulations are suspensions formed by complexing the hormone with a protein that slowly releases the insulin. Onset ranges from 30 minutes to 3 hours, peak effect is 2-16 hours, and duration is up to 24 hours. Long-acting insulins are also complexed; onset is 2-6 hours, peak is 8-24 hours, and duration is up to 36 hours. Actual pharmacodynamics will vary depending on injection site, dose, and patient.
Usually a mixture of short- and intermediate-acting insulins is administered twice daily, one dose before breakfast and the second before the evening meal. Premixed (biphasic) insulins are popular, although some patients prefer to mix their own to better meet individual needs. Once-daily injections of mixed insulins are used primarily by very young or old patients who do not need tight control, and by patients who still produce some insulin but not enough. Multiple injection regimens are popular because of their ability to provide near-normoglycemia, although it is not clear that injections four times daily are better than twice daily. Moreover, repeated injections of short-acting insulin can increase the risk of hypoglycemia or hyperglycemia if a meal or an injection is missed.
Recently Jendle and Karlberg described the hypoglycemic effects of nebulized insulin administered via the lungs to healthy volunteers. Insulin administration through the respiratory tract was first reported in 1925, and subsequently shown to be feasible but not particularly efficacious (only 10% of the amount delivered was retained). However, with the new jet nebulizers, 36% of a dose is retained, making insulin inhalation more practical. In the Jendle study, either saline placebo or human insulin (in 40 U, 80 U, and 160 U doses) was administered to eight volunteers using a new jet nebulizer. At the 160 U dose, there was an increase in mean serum insulin concentration from 0.5 to 26.1 mU/L and reduction in blood glucose concentration from 4.3 to 2.8 mmol/L. Serum peptide-C concentration also dropped by 75%. Insulin absorption across the lung surface peaked at 60 minutes. No adverse reactions were seen and there were no significant changes in lung function. The investigators concluded that studies in diabetic subjects are warranted.