Pharmacokinetics and -dynamics of insulin absorption
Exogenously administered insulin will never be able to exactly mimick the effects of insulin endogenously released by the pancreas. The main reason is that the pancreas releases insulin into the portal vein, so that the insulin passes the liver first. There, more than 50% of insulin is extracted, and as a result hepatic exposure to insulin is high, and peripheral (muscle, fat) exposure to insulin is low. When exogenous insulin is administered -whether i.v., s.c., i.m. or otherwise- it is distributed throughout the circulation, exposing the peripheral organs to relatively high and the liver to relatively low levels of insulin. Another problem is that the pancreas releases insulin in small bouts at short intervals, in direct response to the ambient level of glucose, whereas the dose of exogenously administered insulin thusfar is predetermined. The precise effects over time of the various exogenous insulins are however also dependent on their pharmacokinetic and pharmacodynamic characteristics and the mode of administration.
Studying pharmacokinetics and -dynamics
The traditional method to assess the pharmacokinetics (the plasma levels of insulin over time) and the pharmacodynamics (the resulting effects on glucose over time) of exogenous insulins is the isoglycaemic clamp study. In this, the study insulin is injected and i.v. glucose is infused to maintain a stable level of glycemia. The Glucose Infusion Rate (GIR) is considered to represent the pharmacodynamics of the insulin. When studying other subjects than those with type 1 diabetes -who do not have endogenous insulin secretion- additional steps have to be taken to make sure that (residual) endogenous insulin does not interfere with the interpretation of the results. While the technique has its disadvantages, it usually does provide us with time-action profiles that give us an idea about the (time to) insulin peak, the duration of increased insulin levels, the time to onset of glucose lowering action, the (time to) peak glucose lowering action (tGIRmax, GIRmax), the duration of action, and the Area Under the Curve of the glucose infusion (GIR-AUC) which represents total biological efficacy of the insulin.
Mode of administration
Various factors affect the pharmacokinetics of insulin administered. The first among these is of course the mode of administration.
- Intravenous administration (which is to be avoided with longer acting insulins, but can be used for regular insulin and the short-acting insulin analogues) invariably results in a quick distribution of the insulin, with an instantaneous peak followed by a rapid decline: the plasma half-time (t1/2) of insulin is about 10-12 minutes, so most of the insulin bolus will have dissipated one hour after bolus injection[a]
- Intramuscular injection results in a more rapid time-action profile than subcutaneous injection because of the high vascularity of muscles (albeit not so fast as i.v. injection). In settings of extreme dehydration (such as hyperosmolar coma or diabetic keto-acidosis) the i.m. route is preferable to the s.c. route because skin blood flow becomes compromised.
- Subcutaneous injection is the common mode because this can be performed relatively safely and easily by patients themselves. Time-action profiles given for insulins (and the comparisons between them) are usually based on s.c. administration.
- Inhalation (pulmonary administration) has been pursued for short-acting insulins because of the rich vasculature of the lungs. However, for various reasons (safety, commercial inviability etc.) the only registered pulmonary insulin (ExuberaR) was taken from the market one year after launch.
- Nasal and oral modes of administration are still being explored but have not yet shown any convincing results.
Factors influencing insulin absorption after s.c. administration
Over the years, much research effort has been put into getting grip on the sometimes erratic behaviour of insulin profiles following administration of the same dose in the same patient. From this we have learned that a wide range of factors may influence the s.c. insulin pharmacokinetics. Ideally, the insulin depot is injected close to the capillary network at the base of the s.c. fat tissue. Thus, most of the interfering factors relate to differences in local adipose tissue blood flow and diffusion capacity.
- The site of injection matters. Generally speaking, three regions can be used for injection: the abdominal, deltoid (upper arm) and femoral (upper leg) region. Of these, the abdominal area results in the most rapid absorption of the insulin, and the femoral region in the slowest resorption. While the differences are less important for the rapidly absorbed short-acting insulin analogues, conventional wisdom dictates that when using short-acting insulins (=aiming for rapid action) the abdominal region is preferred; when aiming for prolonged action, the femoral region is preferred.
- When the concentration at which the insulin is formulated is higher, the diffusion gradient is stronger and absorption is more rapid. Nowadays, most human insulins are formulated as U-100 (100 IU of insulin in 1 ml) but other formulations are available such as the old-fashioned U-40, and the modern U-500 developed to accommodate those with very high insulin needs.
- The volume and the dose injected play a role. Because a higher volume has a relatively smaller surface area, diffusion and hence absorption of regular insulin from its s.c. depot has been shown to be slower for larger volumes than for smaller volumes. Thus, when administering more than 40-50 IU of rapid acting insulin, it is usually recommended to split the dose in two injections. Of course, the fact that the higher volume is also more difficult and painful to inject in a single spot is also a reason to split the dose (and this is also the reason why U-500 insulin formulations can be useful in those with high insulin needs). A higher dose usually also prolongs action because of higher residual levels of insulin at each timepoint after injection.
- The depth of injection and the thickness of the subcutaneous fat layer together determine the diffusion distance to the capillary layer. Traditionally the (skinny) patients with type 1 diabetes were advised to use a so-called 'lifted skinfold' technique to avoid accidental intamuscular injection, and much effort was put into developing shorter injection needles. These days however, the reverse is probably more of a problem: obese patients failing to get the insulin deep enough to ensure rapid absorption. The adipose tissue blood flow in obese is less, and a recent experiment by Gagnon-Auger clearly demonstrates that even a rapid acting insulin analogue may have serious delays in time to peak and in duration of action following injection of somewhat higher insulin doses in obese type 2 patients.
- Exercise, local massage and heat exposure also increase the rapidity of absorption, probably by increasing skin blood flow. The latter was demonstrated in a classic experiment where the clamp studies were performed while subjects were in a Finnish sauna.
- Finally, of course, the intrinsic properties of the insulin studied are main determinants, but their respective properties are discussed elsewhere.
Insulin is usually cleared by receptor-mediated uptake and intracellular degradation. The main site of plasma extraction is the liver, with smaller contributions by adipose tissue and muscle.
^ Gagnon-Auger M. et al. Diabetes Care 2502-2507
^ It is conventionally accepted to consider a drug gone after 5 half-times, because the residual plasma levels will then be (1/32)times the original plasma level, so less than 3%. However, in accidental or deliberate overdosing of very high doses, longer periods may be necessary