mEDS:
Humulin is synthesized in a special non-disease-producing special laboratory strain of Escherichia coli bacteria that has been genetically altered by the addition of the gene for human insulin production.
Humulin R consists of zinc-insulin crystals dissolved in a clear fluid. Humulin R has nothing added to change the speed or length of its action. It takes effect rapidly and has a relatively short duration of activity (4 to 12 hours) as compared with other insulins.
Humulin N is a crystalline suspension of human insulin with protamine and zinc providing an intermediate-acting insulin with a slower onset of action and a longer duration of activity (up to 24 hours) than that of regular insulin.
Humulin L is an amorphous and crystalline suspension of human insulin with a slower onset and a longer duration of activity (up to 24 hours) than regular insulin.
Humulin U is a crystalline suspension of human insulin with zinc providing a slower onset and a longer and less intense duration of activity (up to 28 hours) than regular insulin or the intermediate-acting insulins (NPH and Lente).
Humulin 50/50 is a mixture of 50% Human Insulin Isophane Suspension and 50% Human Insulin Injection. It is an intermediate-acting insulin combined with the more rapid onset of action than regular insulin. The duration of activity may last up to 24 hours following injection.
Humulin 70/30 is a mixture of 70% Human Insulin Isophane Suspension and 30% Human Insulin Injection. It is an intermediate-acting insulin combined with the more rapid onset of action of regular insulin. The duration of activity may last up to 24 hours following injection.
The time course of action of any insulin may vary considerably in different individuals or at different times in the same individual. As with all insulin preparation, the duration of action of all forms of Humulin is dependent on dose, site of injection, blood supply, temperature, and physical activity. Humulin is a sterile solution and is for subcutaneous injection. It should not be used intravenously or intramuscularly. The concentration of all forms of Humulin is 100 units/ml (U-100).
Rapid-acting insulin
|
|
||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||
|
|
Short-acting insulin (regular)
|
|
||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||
|
|
Intermediate-acting insulin
|
|
||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||
|
|
Long-acting insulin
|
|
|||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||
|
|
Your personal treatment program might include more than one type of insulin to
be used at different times of the day, at the same time each day, or even in
the same injection.
In addition to the insulin types listed above, you can purchase vials,
prefilled pens and cartridges of insulin that have two types of insulin already
mixed together in a set proportion.
Pre-mixed insulins
|
|
||||||||||||||||||||||||||||||
|
With the introduction of several new insulins since 1996, and more on the way, insulin therapy options for type 1 and type 2 diabetes have expanded. Insulin therapies are now able to more closely mimic physiologic insulin secretion and thus achieve better glycemic control in patients with diabetes. This chapter reviews the pharmacology of insulins (using a comparative approach), types of insulin regimens and therapeutic adjustment of them, and provides an overview of insulin pump therapy.
In 1922, Canadian researchers were the first to demonstrate a physiologic response to injected animal insulin in a patient with type 1 diabetes. In 1955, insulin was the first protein to be fully sequenced. The insulin molecule consists of 51 amino acids arranged in two chains, an A chain (21 amino acids) and B chain (30 amino acids) that are linked by two disulfide bonds.(1) (Figure 1) Proinsulin is the insulin precursor that is first processed in the Golgi apparatus of the beta cell where it is processed and packaged into granules. Proinsulin, a single-chain 86 amino acid peptide, is cleaved into insulin and C-peptide (a connecting peptide); both are secreted in equimolar portions from the beta cell upon stimulation from glucose and other insulin secretagogues. While C-peptide has no known physiologic function, it can be measured and if present, indicates a person has functioning beta-cells.
|
|
|
Figure 1: Insulin Structure |
Insulin exerts effect on glucose metabolism by binding to insulin receptors throughout the body. Upon binding, insulin promotes the cellular uptake of glucose into fat and skeletal muscle and inhibits hepatic glucose output, thus lowering the blood glucose. (HOLD FOR HYPERLINK TO SEE OTHER INSULIN CHAPTER FOR DETAILS)
Commercially available insulins are used for all people with type 1 diabetes, in whom insulin is required for survival and for people with type 2 diabetes when diet/exercise and oral agents are no longer enough to provide adequate glucose control.
With the availability of human insulin by recombinant DNA technology in the 1980's, use of animal insulin declined dramatically. Beef insulin and beef-pork insulin are no longer commercially available. The FDA may allow for personal importation of beef insulin from a foreign country if a patient cannot be treated with human or pork insulin. (2) Pork insulin is still available in the US (Iletin II, Lilly). Beef insulin differs from human insulin by 3 amino acids and pork insulin differs by one amino acid. (1) Eli Lilly discontinued marketing the combination beef-pork insulin (Iletin I). The structure of the insulin molecule is shown in Figure 1 and indicates the amino acid differences among human, beef and pork insulin.
|
Insulin Amino Acids |
|||
|
|
A-Chain |
B-Chain |
|
|
Source |
Position 8 |
Position 10 |
Position 30 |
|
Beef |
Alanine |
Valine |
Alanine |
|
Pork |
Threonine |
Isoleucine |
Alanine |
|
Human |
Threonine |
Isoleucine |
Threonine |
Currently, most insulin used is ether human insulin or and analog of human insulin. The recombinant DNA technique for human insulin involves insertion of the human proinsulin gene into either Saccharomyces cerevisiae (baker's yeast) or a non-pathogenic laboratory strain of Escherichia coli (E coli) which serve as the production organism. Human insulin is then isolated and purified. (3-7)
Recombinant DNA technology has allowed for the development and production of analogs to human insulin. With analogs, the insulin molecule structure has been modified slightly to alter the pharmacokinetics properties of the insulin, primarily affecting the absorption. The B26-B30 region of the insulin molecule is not critical for insulin receptor recognition and it is in this region that amino acids have been substituted. (8) Thus, the insulin analogs are still recognized by and bind to the insulin receptor. The structures of the three insulin analogs currently available are shown in Figure 2 (insulin aspart and lispro) and Figure 3 (insulin glargine).
|
|
|
Figure 2: Insulin Aspart and Lispro Structure |
|
|
|
Figure 3: Insulin Glargine (LANTUS) Structure |
Because insulin analogs are modified human insulin, the safety and efficacy profiles of these insulins have been compared to human insulin. (8) Insulin and IGF-1 receptor binding affinities (IGF- insulin like growth factor), metabolic and mitogenic potencies of insulin lispro, insulin aspart, and insulin glargine relative to human insulin has been assessed. Insulin lispro and aspart are similar to human insulin on all of the above parameters, except insulin lispro was found to be 1.5-fold more potent in binding to the IGF-1 receptor compared to human insulin. Insulin glargine was found to have a 6- to 8-fold increase in mitogenic potency and IGF-1 receptor affinity compared to human insulin. While the clinical significance of these differences is not known, they likely do not represent any significant concern. (9)
Because pork and beef insulin differ from human insulin by 1 and 3 amino acids respectively, they are more immunogenic than exogenous human insulin. Older formulations of insulin were less pure, containing islet-cell peptides, proinsulin, C-peptide, pancreatic polypeptides, glucagons, and somastostatin, which contributed to immunogenicity of insulin. (10) Components of insulin preparations (e.g., zinc, protamine) and subcutaneous insulin aggregates are also thought to contribute to antibody formation. (10) Commercially available human insulins are now virtually free of contaminants and contain <1 ppm of proinsulin (also referred to as "purified"). (1) Insulin side effects such as local or systemic hypersensitivity, lipodystrophy, and antibody production causing insulin resistance, are now rarely seen with exogenous human insulin. (1) Because of the availability of human insulin and the increased potential for animal source insulin to be immunogenic, animal source insulins are now rarely used and people with diabetes should be initiated on human insulin.
The rare hypersensitivity responses to insulin can be immediate-type, local or systemic IgE-mediated reactions. (10) Patients who experience a true allergic reaction to insulin often have received insulin in the past, and experience the allergic reaction after insulin is restarted. Another allergic reaction seen with animal insulins is a delayed local reaction that is IgG-mediated. (10) Insulin therapy can also result in the production of insulin antibodies of the IgG class, which neutralize insulin. Immunological insulin resistance can occur in patients with very high titers of IgG-antibodies.
Lipodystrophy seen with insulin can be two conditions: lipoatrophy and lipohypertrophy. Lipoatrophy is an immune-mediated condition in which there is loss of fat at the insulin injection sites. (10) Lipoatrophy occurs much less frequently with purified human insulins. Treatment for patients who were on an animal insulin was injection with human insulin at the atrophied site. Lipohypertrophy is a non-immunological side effect of insulin resulting from repeated administration of insulin at the same injection site.
In the United States, all insulins are available in the concentration of 100 units/ml (signified as U-100). Insulin syringes are designed to accommodate this concentration of insulin. Regular human insulin (Humulin R, Lilly) is available in a more concentrated insulin, U-500 (500 units/ml), however this preparation is used primarily in a specialized institutional setting or for rare cases of extreme insulin resistance, where very large doses of insulin are required.
Outside the United States, a less concentrated insulin preparation, U-40, (40 units/ml) is still available and sometimes used. Specific U-40 syringes are used with this insulin. It is important that patients traveling from one country to the next, be aware of the concentration of insulin they use, and that the appropriate syringe is used.
Physical and Chemical Properties
Regular human insulin is crystalline zinc insulin dissolved in a clear solution. It may be administered by any parenteral route: subcutaneous, intramuscular, or intravenous. Insulin lispro and aspart are also soluble crystalline zinc insulin, but are intended for subcutaneous injection. NPH, lente and ultralente insulins are suspensions with the regular insulin in them either complexed with protamine (NPH, Neutral Protamine Hagedorn, or isophane, meaning stoichiometric proportions) or mixed with excess amounts of zinc ions (lente and ultralente) to delay their absorption and thus extend their actions. (1,11) Insulin suspensions should not be administered intravenously. All insulins, except insulin glargine, are formulated to a neutral pH.
Insulin glargine is a soluble clear insulin and has a pH of 4.0. Its acidic pH is critical for its subcutaneous (SC) absorption characteristics and will be discussed further under pharmacokinetics. Insulin glargine should only be administered subcutaneously. (5)
Absorption
Insulin administered via SC injection is absorbed directly into the bloodstream, with the lymphatic system playing a minor role in transport. (1) The absorption of human insulin after SC absorption is the rate limiting step of insulin activity. This absorption is inconsistent with the coefficients of variation of T50% (time for 50% of the insulin dose to be absorbed) varying ~25% within an individual and up to 50% between patients. (1,12) Most of this variability of insulin absorption is correlated to blood flow differences at the various sites of injection (abdomen, deltoid, gluteus, and thigh). (1) For regular insulin, the impact of this is a ~2 times faster rate of absorption from the abdomen than other subcutaneous sites. (1) The clinical significance of this is that patients should avoid random use of different body regions for their injections. For example, if a patient prefers to use their thigh for a noontime injection, this site should be used consistently for this injection. For simplicity, however, the abdomen is often recommended as the preferred site of injection because it is the least susceptible to factors affecting insulin absorption (see Table 1). Insulin lispro and insulin aspart appear to have less day-to-day variation in absorption rates and also less absorption variation from the different body regions. (3, 6, 14) Insulin glargine's pharmacokinetic profile is similar after abdominal, deltoid or thigh SC administration. (5)
|
Table 1. Factors Affecting Insulin Absorption |
|
|
Factor |
Comment |
|
Exercise of injected area |
Strenuous exercise of a limb within 1 hour of injection. Clinically significant for regular insulin |
|
Local massage |
While its OK to press on the injection site to prevent seepage, site should not be rubbed vigorously or massaged |
|
Temperature |
Heat
can |
|
Site of injection |
Insulin is absorbed faster from the abdomen. Less clinically relevant with rapid-acting insulins and insulin glargine |
|
Lipohypertrophy |
Injection into areas with this delays insulin absorption |
|
Jet injectors |
|
|
Insulin mixtures |
Loss of short-acting action of regular insulin when mixed with Lente; inadequate resuspension of suspension insulins. |
|
Insulin dose |
Larger doses have delay in action and duration. |
|
Physical status (soluble vs. suspension) |
Suspension insulins must be sufficiently resuspended prior to injection to reduce variability |
|
Modified from References 1, 12, 13. |
|
absorption
rate. Avoid the sauna, shower, hot bath soon after injection. Cold has the
opposite effect.A general principle for factors that can alter insulin absorption is that when local blood flow in the SC tissue is changed, the absorption rate of insulin will also be affected. A factor that increases SC blood flow will increase the absorption rate and vice versa. See Table 1 for factors that affect insulin absorption.
Distribution
Circulating insulin is distributed in equilibrium between free insulin and insulin bound to IgG antibodies.(1) The presence of insulin antibodies can delay the onset of insulin activity, reduce the peak concentration of free insulin, and prolong the biologic half-life of insulin. (15)
Elimination
The kidneys and liver account for the majority of insulin degradation. Normally, the liver degrades ~60% of insulin released by the pancreas (insulin delivered through portal vein blood flow) and the kidneys ~35-45%. (14) When insulin is injected exogenously, the degradation profile is altered since insulin is no longer delivered directly to the portal vein. The kidney has a greater role in insulin degradation with SC insulin (~60%), with the liver degrading ~30-40%. (16)
Because the kidneys are involved in the degradation of insulin, renal dysfunction will reduce the clearance of insulin and prolong its effect. This decreased clearance is seen with both endogenous insulin production (either normal production or that stimulated by oral agents) and exogenous insulin administration. Renal function generally needs to be greatly diminished before this becomes clinically significant. (17)
The onset, peak, and duration of effect are the most clinically significant differences among the insulins. Insulin pharmacodynamics refers to the metabolic effect of insulin. Commercially available insulins can be categorized as rapid-acting (or ultra-short-acting), short-acting, intermediate-acting, and long-acting. The current insulins available in the US are listed in Table 2. Insulin pharmacodynamics, ( onset, peak and duration of the various insulins) are shown in Table 3. It is important to note that ranges are listed for the onset, peak and duration, accounting for intra/inter-patient variability. Each patient will have an individual pattern of response. By having the patient self-monitor their blood glucose frequently, the patient-specific time-action profile of the specific insulin can be better appreciated. Figures 4(a-e),5 graphically show the time-activity profiles for the various insulins.
|
Table 2. Insulins Commercially Available in the US |
|||
|
Category/ |
Source |
Brand Name |
Preparation(s) |
|
Rapid-Acting |
|||
|
Insulin Lispro |
Recombinant DNA |
Humalog (Lilly) |
vial, cartridge, disposable pen |
|
Insulin Aspart |
Recombinant DNA |
NovoLog (Novo Nordisk) |
vial, cartridge |
|
Short-Acting |
|||
|
Regular |
|||
|
Purified |
Pork |
Iletin I Regular (Lilly) |
vial |
|
Human |
Recombinant DNA |
Humulin R (Lilly) |
vial, cartridge, disposable pen |
|
|
|
Novolin R (Novo Nordisk) |
vial, cartridge |
|
|
|
Velosulin BR (Lilly) |
vial |
|
Intermediate-Acting |
|||
|
NPH |
|||
|
Purified |
Pork |
NPH Iletin II (Lilly) |
vial |
|
Human |
Recombinant DNA |
Humulin N (Lilly) |
vial, cartridge, disposable pen |
|
|
|
Novolin N (Novo Nordisk) |
vial, cartridge, disposable pen |
|
Lente |
|||
|
Purified |
Pork |
Lente Iletin II (Lilly) |
vial |
|
Human |
Recombinant DNA |
Humulin L (Lilly) |
vial, cartridge, disposable pen |
|
|
|
Novolin L (Novo Nordisk) |
vial, cartridge, disposable pen |
|
Long-Acting |
|||
|
Ultralente |
Recombinant DNA |
Humulin U Ultralente (Lilly) |
vial |
|
Insulin Glargine (LANTUS) |
Recombinant DNA |
Lantus (Aventis) |
vial |
|
Insulin Mixtures |
|||
|
NPH/Regular (70%/30%) Human |
Recombinant DNA |
Humulin 70/30 (Lilly) |
vial, cartridge, disposable pen |
|
|
|
Novolin 70/30 (Novo Nordisk) |
vial, cartridge, disposable pen |
|
NPH/Regular (50%/50%) |
Recombinant DNA |
Humulin 50/50 (Lilly) |
vial |
|
NPL/Lispro (75%/25%) |
Recombinant DNA |
Humalog Mix 75/25 (Lilly) |
disposable pen |
|
Aspart Protamine/Aspart (70%/30%) |
Recombinant DNA |
NovoLog Mix 70/30 |
cartridge, disposable pen |
|
Note: Insulin analogs are available by prescription only. Human and pork insulin is available over-the-counter. |
|||
|
Table 3. Insulin Pharmacodynamics |
||||
|
Insulin |
Onset (hr) |
Peak (hr) |
Duration (hr) |
Appearance |
|
Insulin Lispro |
within 15 min |
½-1½ |
3-5 |
clear |
|
Insulin Aspart |
within 15 min |
1-3 |
3-5 |
clear |
|
Regular |
½-1 |
2-4 |
5-8 |
clear |
|
NPH |
1-2 |
4-10 |
14+ |
cloudy |
|
Lente |
1-3 |
6-14 |
20+ |
cloudy |
|
Ultralente |
6 |
14-18 |
18-24 |
cloudy |
|
Insulin Glargine |
1.5 |
flat |
24 |
clear |
|
Source:
References 3,5,6,12,13, 18-22 |
||||
|
|
|
Figure 4a. |
|
|
|
Figure 4b. |
|
|
|
Figure 4c. |
|
|
|
Figure 4d. |
|
|
|
Figure 4e. |
|
|
|
Figure 4f. |
|
|
|
Figure 5. |
Dose-Dependent Effect
The pharmacodynamics of regular, NPH and lente insulins are particularly affected by the size of the dose.(1) Larger doses can cause a delay in the peak and increase the duration of action. For example, injecting 4 units of NPH will have a significantly different time-action profile compared to 30 units of NPH.
Rapid-Acting Insulin
Insulin Lispro
Insulin lispro [Lys (B28), Pro (B29)] is an insulin analog that was approved June 14 1996. The B28 (proline), B29 (lysine) amino acid sequence of the insulin molecule is reversed to be lysine-proline resulting in a rapid absorption, within 15 minutes. Because it is absorbed more rapidly, its onset and peak are sooner (and duration shorter) compared to regular insulin. Humalog is also approved for injection immediately after a meal. Because insulin lispro can be injected just before (or after) the meal versus waiting 30 minutes with regular insulin, patients may find it provides them with more flexibility and convenience for their mealtime insulin injection. Insulin lispro can be more effective in lowering postprandial blood glucose levels and has a reduced risk of hypoglycemia compared to regular insulin. (26-28) The reason insulin lispro is associated with less hypoglycemia is due to better matching of insulin effect and food absorption. (28) While Humalog has been studied for use in insulin pumps, it does not have FDA approval for this indication. (29-31) In the rare case of severe insulin allergy, insulin lispro has been shown to be less immunogenic. (32)
Insulin Aspart
Insulin aspart is a human insulin analog approved June 7, 2000. The B28 amino acid proline is substituted with aspartic acid resulting in a rapid onset of activity. Insulin aspart should be injected 5-10 minutes before the meal. Advantages listed above for insulin lispro are the same for insulin aspart. (33)
The insulin aspart preparation Novolog is approved for use in insulin pumps.
While on a molar basis insulin aspart and lispro have identical in vivo potency compared to regular human insulin, higher peak concentrations are achieved. (14) Thus, while a 1:1 conversion is often used for the initial switch from regular insulin to insulin aspart or lispro, over time, a patient's dose of insulin aspart or lispro may need to be lowered. This dosing change is also due to the better matching of the peak of the insulin with the meal, thus achieving better post-prandial control.
Short-Acting Insulins
Regular insulin has an onset of action of 30-60 minutes. It should be injected approximately 30 minutes before the meal. Adherence to this can be inconvenient and difficult for some patients.
Intermediate-Acting Insulins
NPH and lente insulin are intermediate-acting insulins whose pharmacodynamic profiles, are generally similar: onset of action is approximately 2 hours, peak effect at 6-14 hours, and duration of action up to 24 hours (depending on the size of the dose). The difference between NPH and lente insulin is subtle; lente insulin can have a slightly delayed peak compared to NPH. Intermediate-acting insulins can serve a basal insulin and/or prandial insulin depending on their time of administration. (more on this with types of regimens)
Long-Acting Insulins
Long-acting insulins serve to provide a basal (or baseline) level of insulin. While ultralente is stated as having a duration of action of approximately 24 hours, many patients will require twice-daily administration to maintain an adequate level of basal insulin to suppress hepatic glucose output throughout a 24-hour period. Ultralente does have a peak effect (between 14-18 hours), which needs to be taken into account when using this insulin for basal purposes.
Insulin Glargine (LANTUS)
Insulin glargine (21A-Gly-30Ba-L-Arg-30Bb-L-Arg-human insulin) is an insulin analog approved April 20, 2000. It consists of two modifications to human insulin. Two arginines are added to the C-terminus of the B chain shifting the isoelectric point of the insulin from a pH or 5.4 to 6.7. (34) This change makes the insulin more soluble at an acidic pH and insulin glargine is formulated at a pH of 4.0. (34) The second modification is at the A21 position, where asparagine is replaced by glycine. This substitution prevents deamidation and dimerisation that would occur with acid-sensitive asparagine. When insulin glargine is injected into subcutaneous tissue, which is at physiologic pH, the acidic solution is neutralized. Microprecipitates of insulin glargine are formed, from which small amounts of insulin are released throughout a 24-hour period, resulting in a low level of insulin throughout the day.(35) The biological activity of insulin glargine is due to its absorption kinetics and not a different pharmacodynamic activity (e.g., stimulation of peripheral glucose uptake). (36)
It is critical that insulin glargine not be mixed in the same syringe with any another insulin or solution because this will alter its pH and thus affect its absorption profile. Lantus may be given at any time of day. Insulin glargine has been shown to have less nocturnal hypoglycemia when used at bedtime compared with NPH insulin. (37, 38) When switching from NPH or ultralente insulin, a conservative approach is to use approximately 80% of the NPH/ultralente dose as insulin glargine and then titrate (see Adjustments below). (39)
All insulin products have an expiration date. This expiration date labeled on the product (vials, cartridges, pens and other delivery devices) applies when they are unopened and refrigerated. Unopened (i.e., insulin not currently in use) insulin should be stored in the refrigerator at 36°F-46°F (2°C-8°C). Insulin should never be frozen or stored in an ambient temperature greater than 86°F (30°C). An insulin vial in use may be kept at room temperature (59°F-86°F) for ~1 month. Insulin cartridges, prefilled disposable pens and other delivery devices can have different storage recommendations for room temperature (see Table 4). Once opened, insulin cartridges and pens should not be refrigerated.
|
Table 4. Insulin Storage Recommendations at Room Temperature |
||
|
Product Name |
Type |
Room Temperature (59-86°F) |
|
Humulin
(R, NPH, Lente, Ultralente) |
Vial |
28 days |
|
Humulin
Regular |
Cartridge |
28 days |
|
Humulin |
||
|
NPH |
Cartridge |
7 days |
|
NPH |
Pen |
14 days |
|
70/30 |
Cartridge |
7 days |
|
70/30 |
Pen |
10 days |
|
Novolin |
||
|
NPH |
Cartridge |
14 days |
|
70/30 |
Cartridge |
10 days |
|
Humalog |
Vial |
28 days |
|
Pen |
28 days |
|
|
Novolog |
Vial |
28 days |
|
Flex Pen |
28 days |
|
|
Humalog Mix 75/25 Pen 10 days |
|
|
|
Novolog Mix 70/30 |
Vial |
28 days |
|
FlexPen |
14 days |
|
|
Cartridge |
14 days |
|
|
Lantus |
Vial |
28 days |
|
Innolet |
||
|
70/30 |
Delivery device |
10 days |
|
NPH |
14 days |
|
|
R |
28 days |
|
|
Velosulin BR |
Vial |
30 days |
The most significant adverse effect of insulin is hypoglycemia. In the DCCT (Diabetes Control and Complications Trial), intensive insulin therapy was associated with a 2-3 fold increase in severe hypoglycemia (i.e., a person requiring assistance). (40) Likewise, in the UKPDS (United Kingdom Prospective Diabetes Study), insulin therapy in the intensively treated group resulted in 1.8% rate of major hypoglycemic episodes compared to 0.7% in the conventional group. (41) All patients receiving insulin should be aware of the symptoms of hypoglycemia and how to treat it.
Weight gain is another significant side effect of insulin therapy. In part, the weight gain can be a result of frequent hypoglycemic episodes in which patients often overtreat/overeat in response to hunger. Insulin, being an anabolic hormone, also promotes the uptake of fatty acids into adipose tissue. The amount of weight gain in the DCCT and UKPDS associated with insulin therapy was 4.6 kg and 4.0 kg respectively. (40, 41).
True allergic reactions and cutaneous reactions are rare (see Immunogenicity). To avoid lipohypertrophy, patients should be instructed to rotate their insulin injection sites, preferably rotating within one area (e.g., abdomen; avoid 2-inch radius around navel) and not reusing for one week. (42)
Whether two different insulins can be mixed in the same syringe depends on the type of insulins and the duration the insulins are kept in the same syringe. Regular and NPH insulin can be mixed in the same syringe for up to 3 months refrigerated without changing the pharmacodynamic profiles of the insulins. (43) The insulin manufacturers recommend that regular insulin should only be mixed with Lente or Ultralente immediately before injection, otherwise, the excess zinc necessary for the formulation of Lente or Ultralente will bind regular insulin and blunt the effect of regular insulin, converting it to a longer acting insulin. Numerous studies have documented this effect, and some practitioners avoid mixing the lente family insulins with regular all together. (44-47)
Insulin lispro may be mixed with NPH and Utlralente insulin without significantly changing the peak effect or total bioavailability of insulin lispro. (3, 48, 49)
Insulin aspart may be mixed with NPH insulin immediately before injection. While this may cause some attenuation of the peak concentration of insulin aspart, the time to peak and total bioavailability are not significantly affected. (6)
When mixing 2 types of insulins, the clear insulin (excluding insulin glargine) should be drawn first into the syringe, then the cloudy insulin. This avoids any cloudy insulin (e.g., NPH, Lente, or Ultralente) from being inserted into the clear insulin vial (e.g., regular) which could affect the pharmacodynamic properties of the clear insulin.
The clear insulin that cannot be mixed with any other insulin is insulin glargine. Because insulin glargine's absorption profile is due to its low pH (4.0), mixing with other insulins (which have physiologic pH), will alter its absorption. Patients who reuse their syringes should specially label one that is used only for insulin glargine.
Some patients have difficulty mixing insulins in the same syringe which can result in inaccurate doses. (50) Patients should demonstrate their ability to accurately mix their insulin to a practitioner. If they are unable to correctly prepare the insulin dose, they should avoid mixing insulins, or can use a fixed, premixed insulin (if appropriate).
Type 1 Diabetes
With decreasing beta cell function resulting in decreased insulin production, people with type 1 diabetes may require insulin for survival. In general, insulinopenic type 1 diabetics generally require 0.5-1.0 units per kg of body weight per day of insulin. (51) Insulin therapy is often initiated at 0.5-0.75 units/kg/day. (51) During the early stages of type 1 diabetes, patients will require less insulin because the beta cells are still producing some insulin; insulin requirements can be in the range of 0.1-0.6 units per kg per day. (52, 53) Intensive insulin therapy (defined as ³3 insulin injections daily) is indicated for people with type 1 diabetes as this has been shown to provide better glycemic control than 1 or 2 daily injections and reduce the development and progression of microvascular complications. (40)
Type 2 Diabetes
Many patients with type 2 diabetes will eventually require insulin therapy. Since type 2 diabetes is associated with insulin resistance, insulin requirements can exceed 1 unit/kg/day. In the UKPDS, by 9 years less than 25% of patients treated with a sulfonylurea as monotherapy were able to maintain A1C levels <7.0%; the majority of patients required insulin therapy within 9 years of diagnosis. (54) When initiating insulin therapy in patients with type 2 diabetes, insulin is used in combination with the oral medications a patient is taking. Often an intermediate to long-acting insulin (e.g., NPH or insulin glargine) is added at bedtime or 70/30 insulin before dinner. (55) The rationale is that insulin, by suppressing hepatic glucose output during the night, will control the fasting blood glucose (FPG), while the oral medication(s) continues to control prandial glucose levels and glucose throughout the day. (55) Typically, a starting dose of 10 units is utilized, or ~0.1-0.2 units/kg. (39). The intermediate to long-acting insulin is titrated to achieve the FPG target. (see Adjustments below). If the patient has poor glycemic control during the day, daytime insulin can initiated; twice-daily regimen of insulin or multiple daily injections can be used. At this point, the patient is experiencing beta-cell failure. If the patient is taking an insulin secretagogue (e.g., glyburide, repaglinide, etc), it should be discontinued, as insulin will now be replaced exogenously. However, the insulin sensitizing oral agents (i.e., metformin and thiazolidinediones) should be continued) Another option is to discontinue the insulin secretagogue when insulin therapy is started; to avoid potential hypoglycemia. (56)
Goals of Therapy
Before starting a patient on insulin, or adjusting their current insulin therapy, it is important to establish glycemic goals tailored to the patient. The American Diabetes Association currently recommends the following glycemic goals: (57)
|
Preprandial plasma glucose |
90-130 mg/dl |
|
Postprandial plasma glucose |
<180 mg/dl |
|
A1C |
<7% |
For example, if a patient's preprandial blood glucose levels have been in the high 200's, an initial goal might be to lower them to 150 mg/dl. Upon achieving this, a lower goal can be set (e.g., 90-130 mg/dl). In the DCCT, retinopathy initially worsened during the first year in patients (with type 1 diabetes) who received intensive therapy. (40) This is thought to be due to rapid lowering of glucose levels. Thus in patients with proliferative retinopathy or those with high A1C (e.g, >10), slower lowering of glucose is warranted. (58) Another example of individualizing glycemic goals, is a patient with hypoglycemic unawareness; glycemic goals should be less aggressive as glucose levels should not border around 70 mg/dl too closely.
Replacement Strategies
Physiologic Insulin Replacement
A functioning pancreas releases insulin continuously, to supply a basal amount to suppress hepatic glucose output between meals and overnight, and also releases a bolus of insulin prandially to promote glucose utilization after eating. (52) Replacing insulin in a manner that attempts to mimic physiologic insulin release is often referred to as the basal/bolus concept. This physiologic replacement requires multiple daily injections (3 or more) or use of an insulin pump. Basal insulin requirements are approximately 50% of the total daily amount. Prandial insulin is ~10-20% of the total daily insulin requirement at each meal. (52) Providing basal-bolus insulin regimens allow patients to have flexibility in their mealtimes and achieve better glycemic control.
Non-Physiologic Insulin Replacement
When insulin is given once or twice daily, insulin levels do not mimic physiologic insulin release patterns. For people with type 2 diabetes, in whom basal insulin replacement is not as critical, once or twice daily regimens can work satisfactorily and reasonable glycemic control achieved.
Examples of Regimens
Once Daily Insulin Regimen (for patients with type 2 diabetes on oral agents)
NPH (figure 6a), ultralente or insulin glargine (LANTUS) (figure 6b) given at bedtime; or for patient who eat large amounts of carbohydrates at dinner, 70/30 insulin can be given prior to dinner (figure 6c)
|
|
|
Figure 6a. |
|
|
|
Figure 6b. |
|
|
|
Figure 6c. |
Twice-daily Insulin Regimen (Split-Mixed and Pre-Mixed Regimens)
Two-thirds of the insulin dose is given in the morning before breakfast and one-third is given before dinner. Premixed insulins can be used or a mixture of a short-acting insulin (e.g., regular, insulin lispro or insulin aspart) and an intermediate-acting insulin (e.g., NPH or lente). (52) (figure 7a)
|
|
|
Figure 7a. |
|
|
|
Figure 7b. |
- 2/3 total daily dose at breakfast: given as 2/3 NPH (or lente) and 1/3 Regular (or insulin lispro/aspart)
- 1/3 total daily dose at dinner: divided in equal amounts of NPH (or lente) and Regular (or insulin lispro/aspart)
For patients who experience nocturnal hypoglycemia when NPH is administered at dinner with a short-acting insulin, moving the NPH dose to bedtime helps reduce the risk for nocturnal hypoglycemia. (59) Conversely, NPH at dinner can result in fasting hyperglycemia due to dissipation of insulin activity and the dawn phenomenon. Moving the NPH dose to bedtime can help resolve this problem. (60) (figure 7b) An obvious limitation to using premixed insulin is reduced flexibility in dosing; if the dose is adjusted, both types of insulin in the mixture are adjusted.
Multiple Daily Insulin Injection Regimen: Basal plus Prandial Insulin
Many different types of regimens are possible with multiple daily injections. Regular, insulin lispro or insulin aspart are used to provide prandial insulin. NPH, lente, ultralente or insulin glargine is used to provide basal insulin.
- Regular, insulin lispro/aspart before meals and NPH, lente, ultralente, or insulin glargine at bedtime. (figure 8a,b)
- Insulin lispro/aspart before meals and NPH, lente, or ultralente twice daily (breakfast and bedtime) (figure 9).
|
|
|
Figure 8a. |
|
|
|
Figure 8b. |
|
|
|
Figure 9. |
Insulin Pumps
Insulin pump or continuous subcutaneous insulin infusion (CSII) therapy is another option for intensive insulin therapy. While pump therapy used to be reserved for primarily type 1 diabetes, patients with type 2 diabetes are now using insulin pumps. Patients initiated on insulin pump therapy need to be very knowledgeable about diabetes management and be practicing self-management. Patients already know how to count carbohydrates and adjust their insulin doses. Potential advantages of insulin pumps include less weight gain, less hypoglycemia, and better control of fasting hyperglycemia due to the dawn phenomenon compared to multiple daily injections. (61-63) (Figure 10)
Timing of Prandial Insulin Injection
The lag time from injecting regular insulin and eating is approximately 30 minutes; while insulin lispro and aspart can be injected within 15 minutes of eating. Depending on the level of hyperglycemia before meals, the lag-time can be increased. Rapid acting insulins allow patients to adjust insulin to match their lifestyle rather than having to adapt the timing of meals to a more fixed insulin regimen.
Adjustments
Insulin doses should be adjusted to achieve glycemic targets. It is always best to err on the conservative side when dosing insulin at initiation or when adjusting current insulin therapy. Typically a 10-20% increase or decrease in an insulin dose is appropriate. If a patient is experiencing hypoglycemia, adjustment of the insulin dose causing the hypoglycemia should be addressed preferentially over other insulin dose adjustments. Hyperglycemia is a domino effect: if a patient is hyperglycemic in the morning, chances are they remain hyperglycemic throughout the day. Therefore, adjust the earliest time of hyperglycemia first. (52)
Adjustment of Intermediate to Long-Acting Insulin:
When a dose of intermediate or long-acting insulin is adjusted, it is recommended to wait at least 2-5 days before further changes in the dose to assess the response. (52)
Adjustment of Once-Daily Evening Insulin
The FPG is used to adjust the intermediate to long-acting insulin given in the evening. A common weekly titration schedule used is: (61)
- If the FPG is >140 mg/dl: Increase by 4 units
- If the FPG is 120-140 mg/dl: Increase by 2 units
For insulin glargine, the following titration schedule has been studied and shown to cause less nocturnal hypoglycemia compared to bedtime NPH insulin. In this study, insulin was titrated, using a forced titration schedule, to target a FPG of £100 mg/dl. (64)
|
Forced Titration Schedule |
|
|
FPG (mg/dL) |
Increase insulin dose |
|
100-120 |
2 |
|
120-140 |
4 |
|
140-180 |
6 |
|
≥ 180 |
8 |
|
Decrease insulin dose (e.g., 2-4 units/day) if hypoglycemia occurs. (modified from reference 64) |
|
Supplemental Insulin for Correction of Hyperglycemia
Regular insulin, insulin lispro or insulin aspart can be used to correct for hyperglycemia. (65) In general, 1-2 units of insulin will lower the blood glucose by 30-50 mg/dl. Often 1 unit for every 50 mg/dl above the glucose target is a starting supplemental dose, adjusting for insulin sensitivity. (61) An example of a supplemental insulin regimen is as follows: For every 50 mg/dl above the premeal glucose target (e.g., 150 mg/dl), add 1 unit of insulin. (66) So, if a person's premeal glucose was 250 mg/dl, 2 units of insulin would be added to the usual dose of premeal insulin. Supplemental insulin can also be used for snacks. (67)
Carbohydrate Counting
A more sophisticated type of insulin regimen is one in which a patient doses their prandial insulin based on the number of carbohydrates eaten at the meal. By learning how to count their carbohydrates, and dosing their insulin accordingly, patients are afforded flexibility in their meals. A starting insulin-to-carbohydrate ration often used is 1 unit of insulin for every 15 grams of carbohydrate. (51) This ratio is adjusted based on insulin sensitivity and may be different for each meal. Carbohydrate counting is too difficult for some patients. In these patients, meal portion sizes and estimates of carbohydrate servings (15 grams) are concepts that can be learned. Medical nutrition therapy is a critical component of therapy for patients on insulin (covered in chapter 13).
A comprehensive diabetes education class, that teaches self-management skills, such as how to dose prandial insulin by matching it to the amount of carbohydrate intake are an excellent resource to facilitate patients in adopting an intensive insulin therapy regimen. (68)
Adjustments for Exercise
Exercise improves insulin sensitivity. Thus, when a patient exercises, it is often necessary to decrease the insulin dose (and increase caloric intake). For morning exercise, the pre-breakfast insulin dose should be reduced (~25% depending on the duration and intensity of the exercise). For late-morning/early-afternoon and evening exercise, the pre-lunch and pre-dinner insulin dose should be reduced respectively. (52) The effect of exercise on insulin sensitivity can last for many hours; so several insulin doses may need to be adjusted.
Self-Monitoring of Blood Glucose
Patients who were not self-monitoring their blood glucose (SMBG) levels prior to insulin need to be educated how to do this, how to interpret their glucose readings, and how to treat hypoglycemia if it occurs. Involvement of diabetes educator is extremely useful when initiating patients on insulin to provide comprehensive self-management training. The ADA currently recommend that people with type 1 diabetes SMBG at least 3 times daily and those with type 2 diabetes at least daily. (57) Most glucose meters are now plasma-referenced, correlating better to the ADA's glycemic goals. Plasma glucose concentrations are approximately 10-15% higher than whole blood glucose concentrations. (69)
A common misconception among patients is that if they are sick enough that they don't eat or even vomit, they should not take their diabetes medications, insulin included. Patients should be instructed to continue their insulin therapy, maintain fluid intake, eat smaller meals as tolerated, and test their glucose levels every 1-4hours (ketones as well for people with type 1 diabetes). Insulin therapy should be adjusted based on the glucose levels. If the glucose is >240 mg/dl with moderate to large ketonuria, patients should contact their provider immediately. (52)
Dietary modification is universally recognized by caregivers as an initial intervention and mainstay for the treatment of overweight patients with T2D (28-30). The other side of the non-pharmacological intervention 'coin' involves changes in lifestyle, usually consisting of increased physical activity (31), smoking cessation, and reduced intake of alcohol. The overall objectives of these approaches are 1) weight loss and exercise training both resulting in improved insulin action, 2) improved glycemic and lipid control (both short-term and chronic), 3) reduced likelihood of developing microvascular and macrovascular complications, and 4) improved quality of life. It is important to remember that patients do not have to achieve their ideal body weight to reap significant health benefit. It has been reported that a loss of 10-20 lb (4.5-9 kg) will be helpful, as long as the weight loss and exercise programs are maintained (32-34).
The recommendations of the American Diabetes Association and the American Association of Clinical Endocrinologists for the nutritional management of patients with T2D are provided elsewhere (35;36) and will not be presented herein. The key to successful dietary intervention is to ensure that caloric intake is less than caloric output. For long-term success, this should be coupled with education designed to improve the patient's understanding of the beneficial effects on blood glucose, lipids, blood pressure, and quality of life. Some successful dietary approaches have included those designed to limit intake of saturated fats (7-10% energy intake), and spreading the nutrient load. Dietary factors that are able to spread the nutrient load include increased frequency of food intake, increased intake of soluble fiber, and increased intake of foods with a low glycemic index. The beneficial role that supplements and foodstuffs high in soluble fiber play in the conventional dietary management of patients with T2D is well documented (37-40).
The average daily intake of carbohydrates recommended for patients with diabetes is approximately 55-60% of total caloric intake, while intake of fat should be limited to approximately 30%. In patients with dyslipidemia, a special effort should be made to limit saturated fat and to substitute unsaturated or monounsaturated fat, especially omega-3 fatty acids. Protein intake should be limited to 10-20% of total caloric intake, and reduced to 10-15% at the onset of macroalbuminuria. With regard to alcohol intake, patients with diabetes are advised to follow-the same precautions as the general population.
Botanicals have been used for medicinal purposes since the dawn of civilization (41). It is well documented (12) that many pharmaceuticals commonly used today are structurally derived from natural compounds found in traditional medicinal plants. The development of the anti-hyperglycemic drug metformin (dimethlybiguanide; Glucophage®) can be traced to the traditional use of Galega officinalis to treat diabetes, and the subsequent search to identify active compounds with reduced toxicity (42-44). G. officinalis is far from the only botanical to have been used as a treatment for diabetes. Chinese medical books written as early as 3000 B.C. spoke of diabetes and described therapies for this disease (45;46). These historical accounts reveal that T2D existed long ago, and medicinal plants have been used for many millennia to treat this disease. To date, the anti-diabetic activities of well over 1200 traditional plants has been reported, although scant few have been subjected to rigorous scientific evaluation for safety and efficacy in humans (45;47-50). This section will provide a brief overview of those botanicals used for diabetes that have received the most scientific attention, and appear worthy to this author of a comprehensive evaluation.
Mormordica charantia
Mormordica charantia is reported to be the most popular plant used worldwide to treat diabetes (48;49). It has many names depending on the geographic location of origin: in India, where it is widely used for diabetes (11), M. charantia is known as karela, bitter melon, and bitter gourd. In other parts of the world, it is also known as wild cucumber, ampalaya, and cundeamor (9). The glucose-lowering activity of M. charantia (administered as both fresh juice and unripe fruit) has been well documented in animal models of diabetes (11;51). Compounds possessing anti-diabetic activity include charantin and vicine (24;49). In addition, other bioactive components of bitter melon extract appear to have structural similarities to animal insulin (51). Several modes of action have been proposed to account for the anti-diabetic activity of M. charantia including inhibition of glucose absorption in the gut, stimulation of insulin secretion, and the stimulation of hepatic glycogen synthesis (9;51). Four clinical trials have reported bitter melon juice, fruit, and dried powder to have a moderate hypoglycemic effect (9;24;52). However, these studies were small and were not randomized or double-blind, and of insufficient quality to recommend the use of M. charantia without careful monitoring. When used at typical doses, 300-600 mg of juice extract or 1-2 g of powdered leaf daily, M. charantia is generally well tolerated, although it should not be used by children or pregnant women (9;52).
Trigonella foenum-graecum
Trigonella foenum-graecum, also known as fenugreek, is an herb native to southeastern Europe, northern Africa, and western Asia, but is also widely cultivated in other parts of the world (53). Fenugreek has a long history of traditional use in both Ayurvedic [holistic system of healing which originated among the Brahmin (Hindu priestly caste) sages of ancient India and Nepal approximately 3000 - 5000 years ago] and Chinese medicine (54), and has been widely used for the treatment of diabetes (49). The defatted seeds of the fenugreek plant contain ~50% fiber (similar to guar gum), along with a variety of bioactive saponins, alkaloids, coumarins, and 4-hydroxyisoleucine, the principle bioactive compound (48;49). This latter compound exhibits insulinotropic activity (55-57); following a 6-day sub-chronic administration, 4-hydroxyisoleucine (50 mg/kg/day) reduced fasting hyperglycemia, insulinemia, and improved glucose tolerance in diabetic rats (57).
The clinical studies that have evaluated the efficacy of fenugreek in individuals with both type 1 and T2D have recently been reviewed (9). In one study, 17 of 21 patients with T2D showed a reduction in 2-hour post-prandial glucose averaging 30 mg/dl following administration of 15 g of ground fenugreek seed (58). In a crossover, placebo-controlled trial with 60 individuals with type 2 diabetes, the treatment group received 12.5 mg defatted fenugreek at lunch and dinner with isocaloric diets for 24 weeks (59). Fasting blood glucose was 151 mg/dl at baseline, and reduced to 112 mg/dl after 24 weeks (P < 0.05). Fenugreek also caused a significant decrease in the area under the glucose curve by approximately 40%.
In a small study, the effects of fenugreek seeds on glycemic control and insulin resistance in mild-to-moderate T2D was performed using a double-blind, placebo-controlled design (60). Twenty-five newly diagnosed patients with T2D (fasting glucose < 200 mg/dl) were randomly divided into two groups. Group I (n = 12) received 1 g/day of fenugreek seeds and Group II (n = 13) received standard care (dietary control, exercise) plus placebo capsules for two months. After two months, fasting blood glucose and two-hour post-glucose load blood glucose were not significantly different. However, following an oral glucose challenge, the area under the curve (AUC) for blood glucose (2375 ± 574 vs 27597 ± 274) and insulin (2492 ± 2536 vs. 5631 ± 2428) was significantly lower (P < 0.001), as was the HOMA-IR index (112.9 ± 67 vs 92.2 ± 57; P < 0.05) in the treatment group compared to control. Serum triglycerides were decreased and HDL cholesterol increased significantly (both P < 0.05) following fenugreek treatment. Significant reductions in total cholesterol and triglycerides have also been reported in other studies following fenugreek treatment (9).
The doses of fenugreek used in clinical studies have ranged from 2.5 grams to 15 grams daily of the crushed and defatted seeds. Crushing is important in order to release the viscous gel fiber, which presumably contributes to the efficacy of fenugreek. Typical doses of seeds are in the range of 1-3 grams mixed with food and taken at mealtime. The most common side effects are gastrointestinal upset (diarrhea and flatulence), which often can be alleviated by dose-titration. Since the fenugreek fiber might absorb other oral medications, fenugreek should be taken independently (e.g. 1-2 h) of other medications. Due to its ability to lower blood glucose, individuals should monitor their glucose levels carefully if used in combination with insulin or other glucose-lowering agents. Fenugreek can exhibit anti-coagulant activity; it should not be used with other anti-coagulating agents due to the increased risk for bleeding.
Gymnema sylvestre
Gymnema sylvestre (also called gurmar), a woody plant that grows wild in India, has a long history of use in Ayurvedic medicine (11). The leaves of G. sylvestre contain glycosides and the peptide gurmarin (48;49). Other plant constituents include resins, gymnemic acids, saponins, stigmasterol, quercitol, and several amino acid derivatives. A water-soluble acidic fraction called GS4 has been used in most clinical studies (see below). Several modes of action have been proposed to account for the anti-diabetic activity of G. sylvestre including increased glucose uptake and utilization, increased insulin secretion, and increased b-cell number (24).
There have been several small clinical studies in individuals with type 1 and T2D (reviewed in (9)). These studies have reported that G. sylvestre decreases fasting glucose and HbA1c, lowered insulin requirements in individuals with type 1 diabetes, and lowered the dose of anti-hyperglycemic medications in individuals with T2D. G. sylvestre also appears to facilitate endogenous insulin secretion, but it is not a substitute for insulin. There are no data from double-blind, placebo-controlled studies in humans that validate the efficacy of G. sylvestre in type 1 or type 2 diabetes. The extract (~400 mg daily) of G. sylvestre is well tolerated, and no significant side effects have been reported.
Coccinia indica
Coccinia indica is a creeper plant (one that spreads by means of stems that creep) that grows wildly in Bangladesh and in many other parts of the Indian sub-continent. Although C. indica has a long history of use as an antidiabetic treatment in Ayurvedic medicine (11), it has not been subjected to the number of clinical trials that have evaluated M. charantia, fenugreek, or G. sylvestre (based on published reports). Neither the bioactive compounds nor mode of action C. indica have been well-characterized, but there is some suggestion that a component(s) of the plant possesses insulin-mimetic activity (24).
A double blind, placebo-controlled trial in which a preparation from the leaves of the plant was administered to patients with uncontrolled T2D for 6 weeks (61). Of the 16 patients who received the experimental preparations, 10 showed significant improvement in their glucose tolerance (P < 0.001), while none out of the 16 patients in the placebo group showed improvement. Several other studies also offered supporting evidence of the beneficial effect of this treatment (reviewed in (24)). No adverse effects were reported. The potential clinical efficacy of this plant warrants further study.
Opuntia (fuliginosa, streptacantha)
Nopal, a member of the Opuntia genus, is widely used in Mexico as a treatment for glucose control (62). It grows in arid regions throughout the Western hemisphere, and is know in the United States as prickly pear cactus. This plant produces both a vegetable, called nopal, and a red egg-shaped fruit called tuna. Nopal is a common component of every day foods including soups, salads, sandwiches, and blended in drinks. When used for glucose control, nopal is prepared as a food, and is available as in bulk, dried powder form or in capsules. The glucose-lowering activity of nopal is likely due to its very high soluble fiber and pectin content (48;49;63), although its ability to reduce fasting glucose is suggestive of additional modes of action (62).
The results of most human studies of this plant have been reported in Spanish-language journals (24;62); two studies evaluating the acute effects of nopal have been published in English by Frati et al (64;65). Both studies used Opuntia streptacantha Lemaire, and reported improved glycemic control (decreased serum glucose) and improved insulin sensitivity (decreased serum insulin) following a single-dose (500 g of broiled or grilled nopal stems) in patients with type 2 diabetes (n= 14 and n = 22). No effect was observed in healthy individuals (64), nor were any adverse effects reported. The potential clinical efficacy of this plant warrants further study.
Panax (ginseng, quinquefolius, japoncicus)
Ginseng, a member of the plant family Aaraliaceae, has been used in traditional Chinese medicine for thousands of years (41;46). The botanical names for Asian ginseng (Chinese or Korean) is Panax ginseng; Japanese ginseng is known as Panax japonicus; American ginseng is Panax quinquefolius. Siberian ginseng belongs to the genus Eleutherococcus. Many therapeutic claims have been ascribed to the use of ginseng root extract including improved vitality and immune function, along with beneficial effects on cancer, diabetes, cardiovascular disease, and sexual function (66). Bioactive compounds that have been identified in ginseng species include ginsenosides, polysaccharides, peptides, and fatty acids (48;49). Most pharmacological actions are attributed to the ginsenosides, a family of steroids named saponins (48;49). A comprehensive review of the randomized controlled trials (RCTs) evaluating ginseng extracts (mostly Panax ginseng and Panax quinquefolius) has been performed, and it was concluded by the authors that there was insufficient evidence to support efficacy for any of the above indications (66).
Two small studies have evaluated the use of ginseng in diabetes. In a double-blind, placebo controlled study, 36 patients with T2D were treated for 8 weeks with ginseng extract (species not specified) at 100 (n =12) or 200 (n = 12) mg/day, or with a placebo (n =12) (67). Ginseng (100 and 200 mg/day) but not placebo lowered fasting blood glucose by approximately 0.5-1 mmol/l (P < 0.05). Eight subjects who were given ginseng and two who were given placebo achieved normal fasting blood glucose. In response to an oral glucose challenge, the area under the 2-h blood glucose curve was reduced approximately 16% (P < 0.001) in the eight ginseng-treated patients who had normalized fasting blood glucose, without any concomitant change in immunoreactive insulin or C-peptide. The 200 mg dose improved HbA1c (~0.5% decrease; P < 0.05) and physical activity compared to placebo. Ginseng had no effect on plasma lipids. Another small study reported the acute effects of Panax quinquefolius administration (single dose on four separate occasions; 3 g/treatment) or placebo on glucose tolerance in patients with T2D (n = 9) and in non-diabetic subjects (n = 10) (68). In both groups, ginseng caused a significant reduction (P < 0.05) in the area under the blood glucose curve by approximately 20% compared to placebo. The potential clinical efficacy of this plant warrants further study.
Aloe vera
The dried sap of the aloe plant (aloes) is a traditional botanical remedy frequently used to treat dermatitis, burns and to enhance wound healing (69), and one of a variety of plants used for diabetes in India (11) and the Arabian peninsula (70). Its ability to lower the blood glucose was studied in 5 patients with type 2 diabetes (71). Following the ingestion of aloe (one-half a teaspoonful daily for 4-14 weeks), fasting serum glucose level decreased in every patient from a mean of 273 ± 25 (SE) to 151 ± 23 mg/dl (P < 0.05) with no change in body weight. This glucose-lowering activity has been confirmed in two other studies which reported that oral administration of the aloes juice (1 tablespoon twice daily) reduced fasting glucose and triglycerides in subjects with type 2 diabetes both in the absence and presence of concomitant sulfonylurea therapy (72;73). No adverse effects were reported in these studies. Aloe gel also holds the potential for glucose-lowering activity, as it contains glucomannan, a water soluble fiber that reportedly has glucose-lowering and insulin sensitizing activities (38;74). The potential clinical efficacy of this plant warrants further study.
Allium (sativum and cepa)
Allium sativum (garlic) has been used as a medicinal herb by the ancient Sumarians, Egyptians, Greeks, Chinese, Indians, and later the Italians and English (75). The leading Indian ancient medical text, Charaka-Samhita recommended garlic for the treatment of heart disease and arthritis for over many centuries (75). Compounds present in aqueous garlic extract or raw garlic homogenate though to be the principle bioactive components include allicin (allyl 2-propenethiosulfonate or diallyl thiosulfonate), allyl methyl thiosulfonate, 1-propenyl allyl thiosulfonate, and g-L-glutamyl-S-alkyl-L-cysteine.
In modern times, garlic preparations have been widely recognized as agents for prevention and treatment of cardiovascular and other metabolic diseases, atherosclerosis, hyperlipidemia, thrombosis, hypertension and diabetes. Epidemiological evidence indicates an inverse correlation between garlic consumption and the reduced risk of the development of cardiovascular disease (76-78). The efficacy of garlic in cardiovascular diseases has been more evident in non-clinical models, thus prompting a variety of clinical trials (75). Many of these trials have reported beneficial effects of garlic on almost all cardiovascular conditions mentioned above; however, a number of studies have reported no beneficial effects, casting doubt on the reputed health benefits of garlic. These differences could have arisen as a result of methodological shortcomings, the use of different formulations/preparations of garlic, heterogeneity of patient populations, dietary inconsistencies, and different time scales of the studies. The glucose-lowering activity of garlic in humans with type 2 diabetes is not well studied. In the studies that have evaluated garlic, the data have been conflicting (24;75). Thus, the role of garlic in glucose control has yet to be confirmed.
Oxidative stress, resulting primarily from chronic hyperglycemia, is a major cause of the complications of diabetes (79). More recently, there is a growing appreciation of the role of oxidative stress as a mediator of insulin resistance and b-cell dysfunction (80;81). In this context, there are a growing number of studies in humans that have reported beneficial effects of antioxidants on various measures of abnormalities of diabetes (26;27). In addition, it is often reported that individuals with diabetes are deficient in one or more essential micronutrients, and that supplementation often provides a significant improvement. This section will provide a concise overview of those antioxidants, vitamins, minerals, and other nutraceuticals that have received the most attention as potential adjunct treatments for diabetes.
a-Lipoic Acid
a-Lipoic acid (LA) is an eight-carbon fatty acid that is synthesized in trace quantities in organisms ranging from bacteria to man (82-84). LA functions naturally as a cofactor in several mitochondrial enzyme complexes responsible for oxidative glucose metabolism and cellular energy production (85;86). LA has been prescribed in Germany for over thirty years for the treatment of diabetes-induced neuropathy (87-89). Results from several recent controlled clinical studies indicate that this compound is safe, well tolerated, and efficacious (89). It is currently in late-stage clinical trials in the US for a similar indication. LA is commercially available in the US as a nutraceutical (dietary supplement).
In addition to the beneficial effects of LA on diabetes-induced neuropathy, several clinical studies have reported an improvement in insulin sensitivity and whole-body glucose metabolism in patients with type 2 diabetes after continuous intravenous (iv) infusion of LA (90-93). Investigators have reported that a continuous infusion iv of LA substantially increases insulin-mediated glucose disposal (~30-50%) (90;91). Oral administration of LA (enteric-coated tablet) exerts a smaller (~20%) but nonetheless significant effect on insulin sensitivity (94;95). To overcome the abbreviated half-life of LA in plasma, a controlled release formulation of LA (CRLA) has been recently developed (96). The pharmacokinetics, safety, and tolerability of CRLA were evaluated in healthy individuals and in patients with type 2 diabetes, and this agent was found to be safe, well-tolerated, and significantly reduced plasma fructosamine in patients with type 2 diabetes (96). Also, non-controlled release LA recently has been reported to increase insulin mediated glucose disposal in patients with type 2 diabetes (97).
Although the exact mechanism of action of LA is unknown, in vitro data from the laboratories of Rudich and others have indicated that LA pretreatment maintains the intracellular level of reduced glutathione (the major intracellular antioxidant) in the presence of oxidative stress, and blocks the activation of serine kinases that could potentially mediate insulin resistance (98-101). Thus, one potential explanation for the protective effects of LA might be related to its ability to preserve the intracellular redox balance (acting either directly or through other endogenous antioxidants such as glutathione), thereby blocking the activation of inhibitory stress-sensitive serine kinases including IKKb . This stress-sensitive kinase is a crucial regulator of the transcription factor nuclear factor-kB (NF-kB), a major target of hyperglycemia, cytokines, reactive oxygen species, and oxidative stress (102-104). The aberrant regulation of NF-kB is associated with a number of chronic diseases including diabetes and atherosclerosis (102;104). The ability of LA to block the activation of NF-kB is well established in vitro and in vivo (100;105-108).
Recent evidence has linked the activation of NF-kB with insulin resistance (80;109). Activation of IKKb inhibits insulin action. Salicylates, which inhibit IKKb activity and block NF-kB activation (110), restore insulin sensitivity both in vitro and in vivo (111;112). Treatment of nine patients with type 2 diabetes for two weeks with high-dose aspirin (7 g/day) resulted in a significant reduction in hepatic glucose production and fasting hyperglycemia, and increased insulin sensitivity (113). The potential for toxicity associated with such a high dose of salicylate administered chronically precludes consideration of this agent for therapy, but the results support the rationale that IKKb inhibition could be a useful pharmacological approach to increase insulin sensitivity. Furthermore, LA and other agents that interfere with the persistent activation of the NF-kB pathway appear to be promising approaches to increase insulin sensitivity, and perhaps even as treatments for complications of diabetes in which NF-kB activation has been implicated (89;103).
Vitamin C
The normal functions of vascular endothelial tissue include regulation of vasomotor tone, inhibition of platelet activity, and regulation of recruitment of inflammatory cells into the vasculature (114). A damaged endothelium ('endothelial dysfunction') is a key event in the development of diabetic macroangiopathy, and is associated with the oxidative stress-mediated blunting of nitric oxide action (115-117). Endothelial dysfunction has been documented in individuals who are insulin resistant, and in those at risk for developing T2D (116;118;119). Acute treatment with vitamin C improved endothelial function in obese subjects (120), in patients with type 1 and T2D, and in women with gestational diabetes (121-123).
In patients with cardiovascular disease including endothelial dysfunction, both acute (single dose, 2 g) and chronic treatment with vitamin C (30 days, 500 mg/d) reverses the vasomotor defect, as judged by increased flow-mediated dilation of the brachial artery (124;125). All of the above studies involved relatively small populations (< 75) and used acute treatment except one, which was for 30 days (125). Nonetheless, the persistent finding of a beneficial effect of antioxidant treatment on endothelial function (flow-mediated dilation) in individuals with demonstrated endothelial dysfunction is encouraging. It is likely that these results will stimulate additional clinical studies of larger size and longer duration to evaluate the efficacy of vitamin C and perhaps other antioxidants.
In addition to playing a major role in the etiology of diabetic macroangiopathy, endothelial dysfunction could promote insulin resistance (118). It is possible that oxidative stress-mediated blunting of nitric oxide action indirectly affects insulin sensitivity (e.g. reduced peripheral blood flow, increased peroxynitrite formation, others) consequently reducing insulin-stimulated glucose transport in skeletal muscle.
Cigarette smoking impairs endothelial function, and is one of the major risk factors for hypertension, atherosclerosis, and coronary heart disease. The effects of vitamin C (infusion) on insulin sensitivity and endothelial function (measured by flow-mediated dilation of brachial artery; FMD) were evaluated in smokers, non-smokers with impaired glucose tolerance, and non-smokers with normal glucose tolerance (126). Both insulin sensitivity and FMD were blunted in smokers and nonsmokers with IGT, compared with controls. In smokers and in non-smokers with impaired glucose tolerance, vitamin C significantly improved FMD, increased insulin sensitivity, and decreased plasma thiobarbituric acid-reactive substances, an index of oxidative stress. In contrast, vitamin C had no effect on these parameters in non-smokers with normal glucose tolerance. In patients with coronary spastic angina and endothelial dysfunction, vitamin C infusion augmented FMD and increased insulin sensitivity (127). In contrast, vitamin C had no effect in healthy controls.
Vitamin E
Cardiovascular disease is the leading cause or morbidity and mortality in the Western world, and the major macrovascular complication of diabetes (128). It is associated with increased oxidative stress (129), and studies both in vitro and in vivo have provided the rationale for numerous prospective clinical studies evaluating the effects of vitamin E (a-tocopherol) on cardiovascular events in different populations (130;131). A review of these data by Jialal and colleagues has led to the overall conclusion that four of the five major prospective trials have reported a beneficial effect on cardiovascular end-points, including cardiovascular death, nonfatal myocardial infraction, ischemic stroke, peripheral vascular disease, and others (131). The one major study (HOPE Study (132)) that was negative for all end-points, had three limitations (131). It was terminated early due to the overwhelming positive effects of the angiotensin-converting enzyme ramipril, it lacked data on the dietary intake of other antioxidants, and only evaluated synthetic vitamin E (a mixture of tocopherols and tocotrienols) and not a-tocopherol, the most potent and effective tocopherol.
In a study in patients with T2D evaluating the effects of vitamin E on biochemical risk factors for the development of cardiovascular disease, vitamin E treatment significantly reduced low-density lipoprotein oxidizability and soluble cell adhesion molecules (133). Taken together, the evidence suggests a beneficial effect of vitamin E in patients with pre-existing cardiovascular disease, and in those who are at a greater risk for its development.
Oral vitamin E treatment appears to be effective in normalizing abnormalities in retinal hemodynamic, and improving renal function in patients with type 1 diabetes of short (disease) duration (134). Vitamin E was beneficial in those individuals with poorest glycemic control and the most impaired retinal blood flow (134). In a well-controlled study, short-term (4 weeks) supplementation of patients with T2D with persistent micro/macroalbuminuria with both vitamins E and C significantly lowered their urinary albumin excretion rate (135). Four months treatment of patients with T2D with autonomic neuropathy with vitamin E improved the ratio of cardiac sympathetic to parasympathetic tone coincident with lowering of several indices of oxidative stress (136). Interestingly, the study also reported a lowering of glycated hemoglobin, insulin, norepinephrine, and the homeoststatic model assessment index, indicative of increased insulin sensitivity and glycemic control. These data suggest that vitamin E and perhaps other antioxidant supplementation may provide a benefit in the treatment of microvascular complications of diabetes including diabetic retinopathy or nephropathy.
Initial reports of a positive effect of vitamin E on insulin action in insulin resistant patients with T2D were published almost ten years ago (137;138). Twenty-five patients with T2D were treated with vitamin E (d-a-tocopherol; 900 mg/day) or placebo for three months in a double-blind, crossover design (139). There was a trend in the reduction of plasma glucose, along with significant reductions in HbA1c levels (7.8 vs. 7.1), triglycerides, free fatty acids, total cholesterol, low-density lipoprotein cholesterol, and apoprotein B (138). The b-cell response to glucose was unaffected. These intriguing results prompted additional evaluations by Paolisso and colleagues using a more sensitive technique to measure insulin sensitivity, the euglycemic-hyperinsulinemic clamp.
Ten healthy subjects and 15 patients with T2D underwent an oral glucose tolerance test and euglycemic-hyperinsulinemic clamp before and after vitamin E supplementation (900 mg/d for 4 mo) (140). In patients with T2D, vitamin E supplementation significantly increased both whole-body glucose disposal (i.e. insulin sensitivity) by approximately 50%, and non-oxidative glucose disposal by approximately 60%. Vitamin E also improved insulin action in the healthy subjects.
Vitamin E also improved insulin action in elderly people (141). Twenty elderly, non-obese subjects with normal glucose tolerance were submitted to euglycemic-hyperinsulinemic clamp in a double-blind, crossover, and randomized study after 4 months treatment with either vitamin E (900 mg/d) or placebo. Whole-body glucose disposal was significantly potentiated by vitamin E compared to placebo. Furthermore, plasma vitamin E concentrations were correlated with net changes in insulin-stimulated whole-body glucose disposal.
In a 4-week, double-blind, randomized study of vitamin E administration (600 mg/d) versus placebo in 24 hypertensive patients, whole-body glucose disposal was measured by the euglycemic-hyperinsulinemic clamp (142). In hypertensive subjects, vitamin E administration significantly increased whole-body glucose disposal, along with the ratio of reduced glutathione/oxidized glutathione in plasma. Four months treatment of patients with T2D with cardiac autonomic neuropathy with vitamin E lowered of glycated hemoglobin, insulin, norepinephrine, and the homeoststatic model assessment index, indicative of increased insulin sensitivity and improved glycemic control (136).
Niacin
Niacin (nicotinic acid) has been used for many years to reduce elevated cholesterol and triglycerides. In addition, niacin has been shown to decrease cardiovascular events and mortality (143). Some degree of angiographic regression has also being shown with niacin when used with other cholesterol medications. However, the use of niacin for the treatment of dyslipidemia-associated T2D has been limited, due to the adverse effect of high doses on glycemic control. Niacin is a B-vitamin (B-3), but when used in the doses necessary for blood cholesterol control, it should be considered a drug and not a vitamin. Recently, it has been reported that niacin has the potential ability, when given in low doses, to be well tolerated and efficacious. In this study, treatment of individuals with dyslipidemia-associated T2D with extended-release niacin (Niaspan™) led to significantly improved lipid levels and minimal changes in glycemic control (144). The extended-release form was designed to circumvent the bothersome side effects of regular niacin, such as flushing of the skin.
In this 16-week, double-blind, placebo-controlled trial, 148 patients were randomized to placebo (n = 49) or 1000 (n = 45) or 1500 milligrams per day (n = 52) of Niaspan™. About half of the study participants continued taking their prescribed statin drugs for cholesterol lowering during the trial, and 81 percent continued their medications for diabetes. Dose-dependent increases in high-density lipoprotein cholesterol levels (+19% to +24%; P < 0.05) vs. placebo for both niacin dosages) and reductions in triglyceride levels (-13% to -28%; P < 0.05) vs. placebo for the 1500-mg Niaspan™) were observed. Baseline and week 16 values for glycosylated hemoglobin levels were 7.13% and 7.11%, respectively, in the placebo group; 7.28% and 7.35%, respectively, in the 1000-mg Niaspan™ group (P < 0.16 vs. placebo); and 7.2% and 7.5%, respectively, in the 1500-mg Niaspan™ group (P < 0.048 vs placebo). Four patients discontinued participation because of inadequate glucose control. Rates of adverse event rates other than flushing were similar for the niacin and placebo groups. Four patients discontinued participation owing to flushing (including 1 receiving placebo). No hepatotoxic effects or myopathy were observed. The authors concluded that low doses of Niaspan™ (1000 or 1500 mg/d) are a treatment option for dyslipidemia in patients with T2D.
Patients with diabetic dyslipidemia are commonly treated with triglyceride-lowering fibrate drugs, but niacin appears more effective than the fibrates for raising HDL. Since most patients with diabetes will require lipid-lowering therapy, the use of statins to lower LDL cholesterol has become routine therapy for the majority of patients. This study suggests that the addition of extended release low-dose niacin to statin therapy could provide an additional benefit for improvement of blood lipids and lipoproteins in patients with diabetes. However, the impact of niacin on glycemic control will still require regular monitoring.
L-Arginine
L-Arginine is classified as a 'semi-essential' amino acid utilized by all cells (145-147). It plays a critical role in cytoplasmic and nuclear protein synthesis, biosynthesis of other amino acids and derivatives, and in the urea cycle. In this essential biochemical pathway, urea is synthesized from arginine to enable the body to remove excess ammonia, which is toxic to cells. L-arginine is classified as a glucogenic amino acid because it can be metabolized into a-ketoglutarate, and enter the citric acid cycle (Kreb's Cycle). In one of its most important roles, L-arginine serves as a direct precursor for the biosynthesis of NO (148). Although this reaction was originally discovered to occur in endothelial cells, the generation of NO from L-arginine occurs in a variety of other cell types including skeletal muscle (146;149;150). NO is produced endogenously from L-arginine in a complex reaction that is catalyzed by the enzyme nitric oxide synthase (NOS). The other product that is formed in this reaction is citrulline. NO serves as a second messenger to trigger blood vessel dilation and increase blood flow. L-arginine is the only physiological substrate that the NOS enzymes use as a nitrogen donor. Thus, under certain conditions, it may be rate limiting for NO production.
It is well established that aging leads to the deterioration of the vasculature and increased risk for cardiovascular disease (151-153). Circulatory diseases account for considerable morbidity and almost half of all deaths in people over the age of 75 years. A major abnormality of the vasculature present in individuals with type 2 diabetes is endothelial dysfunction, or reduced blood flow capacity. As discussed above, NO is a major regulator of the blood flow. Basal release of NO from the vascular endothelium maintains a constant vasodilating tone. Impaired NO-mediated vasodilatation has been described in hypertension, diabetes, cardiovascular disease, and aging (117;154).
In atherosclerosis, the endothelium has a reduced capacity to produce NO and target cells are relatively insensitive to it (117). The ability of NO to cause vasodilation provides an explanation for the mechanism of action of nitroglycerin, which has been used for over 100 years to treat patients with angina (pain due to inadequate blood flow to the heart) (155). NO is produced following administration of nitroglycerin and other NO donors, such as L-arginine (155;156). In particular, L-arginine is a substrate for NOS, which is responsible for the endothelial production of NO.
Therefore, many investigators have evaluated the usefulness of L-arginine supplementation in animals and in humans in increasing NO production and improving cardiovascular health. The results of these studies have been summarized in several recent books (147;157) and a review (158). Results of oral L-arginine supplementation in hypercholesterolemic animals have consistently shown beneficial effects. L-arginine appears to inhibit the progression of atherosclerotic plaques and preserve endothelial function (158). In addition, L-arginine affects other mediators of atherosclerosis, including circulating inflammatory cells and platelets (158).
On balance, the data in humans have also been positive, although more variable (159-170). This variability is likely due to heterogeneous subject populations with a variety of non-standardized clinical symptoms, small sample sizes, abbreviated duration of treatment, and sub-therapeutic treatment doses. Five of the 17 studies showed no cardiovascular health benefit from oral L-arginine supplementation (158). The remaining 12 studies demonstrated beneficial effects as evidenced by decreased platelet aggregation and adhesion, decreased monocyte adhesion, or improved endothelium-dependent vasodilation (158). Taken together, these studies provide supporting evidence for the idea that treatment with an exogenous NO donor could have a beneficial effect on cardiovascular health.
Coenzyme Q10
Another antioxidant reported to have beneficial effects for diabetes is coenzyme Q10. The effects of orally administered coenzyme Q10 were evaluated in a double-blind, placebo-controlled study of 30 patients with coronary heart disease (171). Following 8 weeks of treatment with coenzyme Q10 (60 mg twice daily), patients exhibited reduced plasma levels of glucose, insulin, (fasting and 2 hour), and lipid peroxides (a marker of oxidative stress) compared to controls. These results indicate that coenzyme Q10 decreased oxidative stress and improved insulin sensitivity.
Chromium
Second only to calcium, chromium is very popular mineral supplement in the US, with over 10 million individual users; in 1999, retail sales of chromium picolinate-containing products totaled over $500 million (172). Several authors, mostly on the basis of small studies of short duration, have suggested dietary trivalent chromium supplementation as an attractive option for the management of T2D and for glycemic control in persons at high risk for T2D (171;173). Thus, chromium has emerged in the US as the most widely used dietary supplement for the treatment of T2D. The link between chromium and carbohydrate metabolism was proposed over 40 years ago, when it was identified as a component of the biologically active 'glucose tolerance factor' (174). Chromium deficiency has been associated with decreased insulin action in both diabetic animals and humans (175-177). A number of human studies have found that chromium supplementation has beneficial effects in individuals with impaired glucose tolerance and diabetes (178). Oral administration of trivalent chromium is associated with favorable safety profile in animals and in humans (179;180).
To critically evaluate the clinical studies with chromium-treatment reported to date, a systematic review and meta-analysis of the RCTs were performed (181). The objective was to determine the effect of chromium on glucose and insulin responses in healthy subjects and in individuals with glucose intolerance or T2D. The authors identified 20 reports of RCTs assessing the effects of chromium on glucose, insulin, or HbA1c. Their analyses summarized data on 618 participants from the 15 trials that reported adequate data: 193 participants had T2D and 425 were in good health or had impaired glucose tolerance. The meta-analysis showed no association between chromium and glucose or insulin concentrations among non-diabetic subjects. A study of 155 diabetic subjects in China reported that chromium reduced glucose and insulin concentrations (182); the combined data from the 38 diabetic subjects in the other studies did not. Three trials reported data on HbA1c: one study each of persons with T2D (182), persons with impaired glucose tolerance (183), and healthy subjects (184). The study of diabetic subjects in China was the only one to report that chromium significantly reduced HbA1c (182). Thus, this meta-analysis of RCTs showed no effect of chromium on glucose or insulin concentrations in non-diabetic subjects, and data for persons with diabetes are inconclusive. Additional RCTs in well-characterized, at-risk populations are necessary to determine the effects of chromium on glucose, insulin, and HbA1c. To this end, the Office of Dietary Supplements (ODS), the National Center for Complementary and Alternative Medicine (NCCAM), and the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) have invited basic and clinical applications to study the role of chromium as adjuvant therapy in T2D and/or impaired glucose tolerance (http://grants1.nih.gov/grants/guide/pa-files/PA-01-114.html).
Vanadium
Vanadium is a transition metal that can exist in several oxidation states (-1 to +5), and is widely present in nature in the form of minerals (185). It is also found in animals and humans, primarily as the tetravalent vanadyl cation (VO2+) and the pentavalent vanadate (VO3-). The tetravalent form is the most common intracellular form whereas the pentavalent form predominates in extracellular body fluids. Animals fed vanadium-deficient diets exhibited an increased rate of spontaneous abortion, depressed milk production, decreased growth, and premature death. The nutritional necessity in humans has not been established. The 'average' diet in the US supplies approximately 15-60 micrograms of vanadium daily. Foods relatively rich in vanadium include black pepper, mushrooms, shellfish, parsley, and dill seed. Fresh fruits, vegetables, and oils contain little or no vanadium.
In vitro and in vivo, vanadium-containing compounds exhibit insulin-mimetic activity primarily due to their ability to inhibit tyrosine phosphatase activity (186;187) and activate a cytosolic tyrosine kinase (188). Many of the metabolic effects of insulin including the stimulation of glucose transport, glycogen synthesis, glucose oxidation, and lipogenesis and anti-lipolysis are mimicked by vanadate and related peroxovanadium compounds (186;189;190). In vivo, vanadate and peroxovanadium compounds significantly lower blood glucose in insulin-dependent and insulin-resistant diabetic animals in the absence of overt toxicity (191-195). The glucose-lowering effect of vanadate is achieved without elevating serum insulin, indicating an insulinomimetic effect and, in some cases, an insulin sensitizing effect.
In humans with T2D, several small studies of 2-4 weeks duration have indicated small but significant beneficial effects of vanadate treatment on various indicators of glucose metabolism (196-201). The most common side effects were gastrointestinal disturbances: nausea, vomiting, diarrhea, and cramps. In the study from Cusi et al (201), eleven patients with T2D were treated with vanadyl sulfate (VS) at a higher dose (150 mg/day) and for a longer period of time (6 weeks) than in the previous studies. Before and after treatment insulin secretion during an oral glucose tolerance test, and hepatic glucose production (HGP) along with whole body insulin-mediated glucose disposal were measured. Treatment significantly improved glycemic control: fasting plasma glucose (FPG) decreased from 194 ± 16 to 155 ± 15 mg/dl, fructosamine decreased from 348 ± 26 to 293 ± 12 mmol/l, and HbA1C decreased from 8.1 ± 0.4 to 7.6 ± 0.4% (all P < 0.01) without any change in body weight. Subjects had an increased rate of HGP compared with non-diabetic controls (4.1 ± 0.2 vs. 2.7 ± 0.2 mg/kg lean body mass/min; P < 0.001), which was closely correlated with FPG (r = 0.56; P < 0.006). Vanadyl sulfate reduced HGP by about 20% (P < 0.01), and the decline in HGP was correlated with the reduction in FPG (r = 0.60; P < 0.05). VS also caused a modest increase in insulin-mediated glucose disposal (from 4.3 ± 0.4 to 5.1 ± 0.6 mg/kg lean body mass/min; P < 0.03), although the improvement in insulin sensitivity did not correlate with the decline in FPG after treatment (r = -0.16; P = NS). Thus, VS at a dose of 150 mg/day for 6 weeks improves hepatic and muscle insulin sensitivity in patients with T2D. The glucose-lowering effect of VS correlated well with the reduction in HGP, but not with insulin-mediated glucose disposal, suggesting that liver, rather than muscle, is the primary target of VS action at therapeutic doses.
Vanadium has a poor therapeutic index, and attempts have been made to reduce the dose of vanadium required for therapeutic effectiveness (185). Organic forms of vanadium, as opposed to the inorganic VS, are appear to be safer (in animal studies), more absorbable, and able to deliver a therapeutic effect up to 50% greater than the inorganic forms (195). An ongoing goal has been to provide vanadium with increased bioavailability, and in a form that is best able to produce the desired biological effects. As a result, numerous organic complexes of vanadium have been developed including bis(maltolato)oxovanadium (BMOV), bis(cysteinamide N-octyl)oxovanadium known as Naglivan, bis(pyrrolidine-N-carbodithioato)oxovanadium, vanadyl-cysteine methyl ester, and bis-glycinato oxovanadium (BGOV) (195;202). The usefulness of these newer formulations as clinical agents remains to be determined. Despite the encouraging results of the clinical studies published to date, the safety of larger doses and use of vanadium salts (or related compounds) for longer periods is unknown (203). Thus, the use of supplemental vanadium for the management of diabetes or impaired glucose tolerance is not recommended at this time.gcl
Magnesium
Magnesium, the fourth most abundant cation in humans, is an essential mineral in human nutrition, and is required for wide array of biological functions. It is a cofactor in over 300 enzymatic reactions, and is important for the electrical stability of cells, maintenance of membrane integrity, muscle contraction, nerve conduction and vascular tone (204). Magnesium deficiency is linked to a number of clinical disorders including insulin resistance, T2D, hypertension, and cardiovascular disease (205-209). Epidemiological studies indicate that high daily magnesium intake is predictive of a lower incidence of T2D (208). The plasma magnesium level is inversely related to insulin sensitivity (210;211), and magnesium supplementation improves insulin sensitivity as well as insulin secretion in patients with T2D (212-214). However, until recently (see below), no beneficial effect of oral magnesium supplementation has been demonstrated on glycemic control either in patients with type 1 or T2D. Nonetheless, RCTs in well-characterized, at-risk populations are warranted to see whether magnesium replacement therapy will prove efficacious in the treatment of T2D.
To this end, it has recently been reported that oral magnesium supplementation (as a solution of magnesium chloride, MgCl2) restores serum magnesium levels, and improves insulin sensitivity and metabolic control in patients with T2D (215). This study was a randomized double-blind placebo-controlled design, in which 63 subjects with decreased serum magnesium (= 0.74 mmol/l) treated by glibenclamide received either 50 ml of MgCl2 solution (50 g/l) or placebo daily for 16 weeks. At the end of the study, MgCl2-treated subjects showed a significantly higher serum magnesium concentration (0.74 ± 0.10 vs. 0.65 ± 0.07 mmol/l, P < 0.02) and lower HOMA-IR index (3.8 ± 1.1 vs. 5.0 ± 1.3, P < 0.005), fasting glucose levels (8.0 ± 2.4 vs. 10.3 ± 2.1 mmol/l, P < 0.01), and HbA1c (8.0 ± 2.4 vs. 10.1 ± 3.3%, P < 0.04) compared to placebo-treated subjects. These results support the use of oral magnesium supplementation in patients with T2D who are magnesium-depleted.
Zinc
Zinc is another essential mineral in human nutrition with a wide range of biological functions. Zinc fulfills catalytic, structural, or regulatory roles in more than 200 zinc-requiring metalloenzymes (204). The interaction of zinc with insulin induces conformational changes and enhances binding to the insulin receptor (216;217). With regard to glucose metabolism, zinc is a co-factor of several key enzymes. Zinc is an activator of fructose-1-6-bisphosphate aldolase, and an inhibitor of fructose-1-6-biphosphatase (218). Zinc can also exert antioxidant activity (219), and is a cofactor in copper/zinc superoxide dismutase, a major antioxidant enzyme (220). Some studies have reported zinc deficiency along with alterations in zinc metabolism in patients with diabetes (218;219). Zinc supplementation studies in patients with diabetes are few, and have yielded contradictory results with regard to effects on glycemic control (218).
Polyunsaturated Fatty Acids
Dietary w-3 polyunsaturated fatty acids (w-3 PUFAs) exhibit a broad array of biological activities in health and disease, including anti-inflammatory, lipid-lowering, and the prevention of coronary heart disease (221-226). The most prominent dietary sources of w-3 PUFAs include fish oils abundant in eicosapentanoic (EPA) and docosahexanoic (DHA) acids along with plants rich in a-linolenic acid. A primary mechanism of action of w-3 PUFAs is achieved by altering gene expression mediated by the regulation of the activities or abundance of four families of transcription factors (227-229). These include the peroxisome proliferator activated receptor (PPAR a,g,d), liver X receptors (a,b), hepatic nuclear factor-4 a, and the sterol regulatory element binding proteins 1 and 2. These transcription factors play major roles in the regulation of hepatic carbohydrate, fatty acid, triglyceride, cholesterol and bile acid metabolism.
A large body of epidemiological and clinical trial data suggests that w-3 PUFAs play a significant role in the prevention of coronary artery disease (223-225). The most convincing evidence is derived from four major intervention trials evaluating either fish meal, fish oil, or an a-linolenic acid-enriched spread on hard clinical end-points including myocardial infarction, death from coronary heart disease, and total mortality (230-233). In essence, these studies found that supplementation significantly reduced cardiovascular events (cardiovascular death, non-fatal myocardial infarction and stoke)(231-233) and total mortality (230). The average recommended intake by an expert panel of US nutritional scientists is 2.2 g/d of a-linolenic acid and 0.65 g/d of EPA plus DHA (234), while the British Nutrition Foundation has recommended 2.4 g/d of a-linolenic acid and 1.2 g/d of EPA plus DHA (235).
Conflicting results have been reported regarding the effects of fish oil supplementation on glycemic control in those with glucose intolerance including individuals with T2D. Several early studies reported detrimental effects (236;237), but subsequent studies with improved design have not replicated the earlier findings (238-241). Results from a meta-analysis of pooled data from all RCTs in which fish oil supplementation was the only intervention in subjects with T2D was recently published (241). Eighteen trials including 823 subjects followed for a mean of 12 weeks were included. Doses of fish oil used ranged from 3 to 18 g/day. The outcomes studied were glycemic control and lipid levels. Meta-analysis demonstrated a statistically significant effect of fish oil on lowering triglycerides (-0.56 mmol/l) and raising LDL cholesterol (0.21 mmol/). No statistically significant effect was observed for fasting glucose, HbA1c, total cholesterol, or HDL cholesterol. The triglyceride-lowering effect and the elevation in LDL cholesterol were most evident in those trials that recruited hypertriglyceridemic subjects and used higher doses of fish oil. Thus, this meta-analysis of RCTs showed that fish oil supplementation in T2D lowers triglycerides, raises LDL cholesterol, and has no statistically significant effect on glycemic control. There is no evidence that fish oil supplementation adversely effects glucose tolerance, insulin action, or insulin secretion in non-diabetic individuals (204).
Conjugated linoleic acid
Conjugated linoleic acid (CLA) refers to a group of polyunsaturated fatty acids that are positional and geometric conjugated dieonic isomers of linoleic acid (242). The biological activity of CLA was originally discovered due to its ability to inhibit chemically induced carcinogenesis in rodents (243;244). Subsequently, numerous health benefits have been attributed to CLA including activity as an anti-obesogenic, anti-diabetogenic, and anti-atherosclerortic agent (245;246). The major isomers of CLA are the cis-9,trans-11 and the trans-10, cis-12, with their biological activities being isomer-specific (242). The major dietary sources of CLA are meat and dairy products. CLA concentrations in dairy products typically range from 2.8 to 7.0 mg/g fat (frozen yogurt and condensed milk, respectively), of which the cis-9,trans-11 isomer comprises ~75%-95% of the total CLA (246). CLA concentrations in meat typically range from 0.6 to 5.8 mg/g fat (pork and lamb, respectively), of which the cis-9,trans-11 isomer comprises ~55%-85% of the total CLA (246). Similar to w-3 PUFAs, CLA isomers are ligands and activators of PPARa, but with an approximate 10-fold higher affinity (~140-260 nM) (247). CLA isomers readily undergo extensive metabolism including elongation and desaturation, yielding additional potential bioactive molecules (245).
In several animal models, CLA has been shown to reduce body fat accumulation, improve glucose tolerance, and increase insulin sensitivity (248;249). In individuals with T2D (n =12), plasma trans-10, cis-12 but not cis-9,trans-11 CLA is inversely correlated with body weight (P < 0.05) and serum leptin (P < 0.02) (250). Thus, CLA supplementation has been suggested as a potential new nutraceutical approach for obesity, a major risk factor for the development of T2D. However, conflicting results have been reported regarding the beneficial effects of CLA supplementation on adiposity and metabolism in humans. In several studies, administration of CLA (1.8-4.2 g/day) for 12 weeks has been reported to decrease body fat mass (~4%; P < 0.001) in healthy individuals (251;252) and in overweight and obese individuals (253). There were no changes in body weight, serum lipids, or glucose metabolism in these studies.
However, in another study, abdominally obese men (n = 60) were treated with 3.4 g/day CLA (isomer mixture), purified trans-10, cis-12 CLA, or placebo (254). Euglycemic-hyperinsulinemic clamp, serum hormones, lipids, and anthropometry were assessed before and after 12 weeks of treatment. Baseline metabolic status was similar between groups. Unexpectedly, trans-10, cis-12 CLA increased insulin resistance (19%; P < 0.01) and glycemia (4%; P < 0.001) and reduced HDL cholesterol (-4%; P < 0.01) compared with placebo. Body fat, sagittal abdominal diameter, and weight decreased versus baseline, but the difference was not significantly different from placebo. The CLA mixture did not change glucose metabolism, body composition, or weight compared with placebo but lowered HDL cholesterol (-2%; P < 0.05). Trans-10, cis-12 CLA also increases markers of oxidative stress and inflammation (255), thus revealing important isomer-specific metabolic actions of CLA in abdominally obese men. In light of these results, the use of supplemental CLA (mixture or individual isomers) for the management of obesity, impaired glucose tolerance, or diabetes is not recommended.
SUMMARY AND POSSIBILITIES FOR TREATMENT
Clearly, many natural products have hypoglycemic, anti-hyperglycemic, insulin sensitizing, anti-hyperlipidemic, anti-hypertensive, and anti-inflammatory activities. There are published studies reporting the anti-diabetic activity of well-over a thousand different botanicals and nutraceuticals. The number of those treatments evaluated in clinical trials is approximately 100 (24). In the vast major of these trials, the botanicals and nutraceuticals were evaluated as an adjunct to diet and prescription medications. Fifty-eight of the trials were controlled, and conducted in individuals with diabetes or impaired glucose tolerance. Of these, statistically significant treatment effects were reported in 88% of trials (23 of 26) evaluating a single botanical, and 67% of trials (18 of 27) evaluating individual vitamin or mineral supplements (reviewed in (24)). When reported, side effects were few and generally mild (gastrointestinal irritation and nausea).
However, many of the studies suffered from design flaws including small (< 10 subjects) sample sizes, heterogeneity of subjects, and short-duration of treatment. Furthermore, there is a lack of multiple studies for many of the individual supplements. Despite the apparent lack of side effects of these treatments, it would be prudent to be aware of the potential for dietary supplements, especially botanicals, to interact with a patient's prescription medication. One of the most important potential botanical-drug interactions is that of garlic, Trigonella, and Ginkgo biloba with non-steroidal anti-inflammatory drugs (including aspirin) or warfarin, as these botanicals possess limited anti-coagulant activity (53;256). Another potential interaction of concern is one involving G. biloba, a botanical widely used for the treatment of memory and concentration problems, confusion, depression, anxiety, dizziness, tinnitus, and headache (257;258). Ingestion of G. biloba extract by patients with T2D may increase the hepatic metabolic clearance rate of not only insulin but also hypoglycemic medications, resulting in reduced insulin-mediated glucose metabolism and elevated blood glucose (259). Another issue to consider with botanicals is the potential for batch-to-batch variation due to age of the plant, geographic source, time of harvest, and method of drying and preparation, all of which can dramatically impact the purity and potency of active ingredients. None of the agents discussed here is recommended for use in pregnant or lactating women, or in children. Furthermore, patients should be advised on the proper use of any alternative treatment to avoid the risk of hypoglycemia.
That being stated, several botanical and nutraceutical agents appear to merit consideration as complimentary approaches for the treatment of T2D. Botanical treatments with the strongest evidence of clinical efficacy include C. indica, T. foenum-graecum, American ginseng, A. vera, and Opuntia (Nopal). Nutraceutical agents with promise for improving insulin sensitivity and glycemic control include a-lipoic acid, vitamins C and E, and magnesium. If the safety profile of vanadium could be confirmed with chronic use, this agent would also be regarded as a promising treatment. In addition, there is evidence that a-lipoic acid and vitamin E improve the symptoms of individuals with microvascular complications including neuropathy and retinopathy. w-3 PUFAs (EPA, DHA, a-linolenic acid), L-arginine, and vitamin C merit consideration for cardiovascular complications.
The increasing movement for the public in general and patients with diabetes (and other diseases) to self-treat using botanicals and nutraceuticals cannot be disputed and should not be ignored. Health care professionals are urged to increase their knowledge base in this area on an ongoing basis. They are also urged to pro-actively query patients on their use of these agents, and record the information obtained in the patient record. Many patients are reluctant to discuss their use of botanicals and nutraceuticals, so it is important for health care professionals to keep an open mind and be non-judgemental. Since patients cannot be expected to distinguish between the marketing hype of manufacturers and evidence derived from credible scientific studies, health care professionals must be positioned to provide an informed opinion and recommendation.
|
Table 1. Major Botanicals Used for Type 2 Diabetes |
||||||
|
Botanical |
Putative Bioactives |
Anti- Diabetic Activity |
Mode of Action |
Typical Daily Dose |
Potential Side Effects |
Relative Rating 9 |
|
M.
charantia 1 |
Charantin, vicine, mormordicine (alkaloid), polypeptide P |
Glucose control |
Inhibits glucose absorption in gut; Enhances insulin secretion; increases glucose transport and glycogen synthesis |
300-600 mg (juice extract); 1.8 g (capsule) |
Hypoglycemia; Additive with other glucose- lowering agents and insulin; GI irritation |
++ |
|
T. foenum- graecum 2 (Fenugreek) |
Fiber, 4- hydroxyisoleucine, saponins, coumarins, alkaloids, glycosides |
Glucose control; Anti- hyperlipidemic |
Delays gastric emptying; Inhibits glucose absorption in gut; Enhances insulin secretion |
2.5-15 g (defatted seeds) |
Hypoglycemia; Additive with other glucose- lowering agents and insulin; GI irritation; Anti- coagulant |
+++ |
|
G. sylvestre 3 (Gurmar) |
Gymnemic acids, gymnemosides |
Glucose control |
Inhibits glucose absorption in gut; Enhances insulin secretion |
200-600 mg |
Hypoglycemia; Additive with other glucose- lowering agents and insulin;GI irritation |
++ |
|
C. indica 4 |
Not yet characterized |
Glucose control |
Not characterized; possibly insulin mimetic |
1.8 g (powdered leaves) |
None reported (limited data) |
+++ |
|
O. streptacantha 5 (Nopal; prickly pear) |
Fiber, pectin |
Glucose control; Anti- hyperlipidemic |
Delays gastric emptying; Inhibits glucose absorption in gut |
2.4 g |
None reported (limited data) |
++ |
|
P. quinquefolius 6 (American ginseng) |
Ginsenosides (saponins), polysaccharides, peptides, fatty acids |
Glucose control |
Delays gastric emptying; Inhibits glucose absorption in gut; Hormonal & CNS activity |
100-200 mg |
Estrogenic effects; Ginseng abuse syndrome; Interacts with many drugs |
++ |
|
A. vera 7 (Aloe) |
Fiber (glucomannan), aloins, anthraquinones, barbaloin, polysaccharides, salicylic acids |
Glucose control |
Not characterized; possibly delays gastric emptying and inhibits glucose absorption in gut |
1.2 g (capsule) |
None reported (limited data) |
++ |
|
A. sativum 8 (Garlic) |
Allicin, allyl methyl thiosulfonate, 1-propenyl allyl thiosulfonate, g-L-glutamyl- S-alkyl-L- cysteine |
Anti-
hyperlipidemic |
Anti- inflammatory (Antioxidant) |
600-1000 mg |
GI irritation; Anti- coagulant; heartburn; garlic odor |
+ |
|
Table 2. Major Nutraceuticals Used for Type 2 Diabetes |
|||||
|
Botanical |
Anti-Diabetic Activity |
Mode of Action |
Typical Daily Dose |
Potential Side Effects |
Relative Rating 14 |
|
a-Lipoic acid 1 |
Insulin sensitizer; Anti- neuropathy |
Anti- inflammatory (Antioxidant) |
900-1800 mg |
GI irritation |
+++ |
|
Vitamin C2 |
Insulin sensitizer; Improves endothelial dysfunction |
Anti- inflammatory (Antioxidant) |
500-2000 mg |
None reported |
++ |
|
Vitamin E3 |
Insulin sensitizer |
Anti- inflammatory (Antioxidant) |
600-900 mg |
None reported |
++ |
|
Niacin4 |
Anti- hyperlipidemic |
Anti- lipolytic; decreases rate of hepatic synthesis of VLDL and LDL |
1000-1500 mg |
Impaired glucose tolerance; flushing |
+++ |
|
L-Arginine 5,13 |
Improves endothelial dysfunction |
Nitric oxide donor |
3-6 g |
None reported |
+++ |
|
Coenzyme Q106 |
Insulin sensitizer |
Anti- inflammatory (Antioxidant) |
100-150 mg |
GI irritation |
+ |
|
Chromium 7 |
Glucose control |
Enhances insulin action |
50-400 mg |
Potential for renal toxicity (rare) |
___ |
|
Vanadium 8 |
Insulin sensitizer; Glucose control |
Insulin mimetic; Pan- tyrosine phosphatase inhibitor |
100-150 mg |
GI irritation; tissue accumulation; Uncertain long-term safety profile |
___ |
|
Magnesium 9 |
Insulin sensitizer; Glucose control Anti- hypertensive |
Not characterized; Possibly enhances insulin action |
300-400 mg; (2.5 g used for glucose control) |
Diarrhea |
++ |
|
Zinc 10 |
Glucose control |
Anti- inflammatory (Antioxidant); insulin binding |
40-80 mg |
GI irritation; metallic taste; headache |
___ |
|
w-3 PUFAs11 |
Anti- hyperlipidemic |
PPARa,g agonists |
0.65-1.2 g (EPA + DHA); 2.4 g ALA |
GI irritation; halitosis |
+++ |
|
CLA 12 |
Anti-obesity |
PPARa agonist |
2-4 g |
GI irritation; Increased inflammation and oxidative stress |
___ |
|
Table 3. Credible Internet Sources for Informationon Botanicals and Nutraceuticals |
||
|
Source |
Website Address |
Thumbnail Sketch |
|
Alternativediabetes.com |
Private site that provides complete and current information about alternative approaches to diabetes treatment, including herbals and nutraceuticals (Under construction as of 6/1/03) |
|
|
The American Society of Pharmacognosy |
Professional organization dedicated to discipline of pharmacognosy (the science and study of drugs from natural sources) |
|
|
Consumer Lab.com |
Independent product review site that provides information on the content of nutritional products including dietary supplements |
|
|
HerbMed |
An interactive, electronic herbal database |
|
|
National Center for Complimentary and Alternative Medicine |
National (US) center that supports and disseminates research results on complementary and alternative medicine |
|
|
Natural Medicines Comprehensive Database |
An extremely comprehensive, scientifically-based, and practical database on natural medicines (Subscription required) |
|
|
NIH Office of Dietary Supplements (ODS) |
National (US) center that supports and disseminates research results on dietary supplements |
|
|
National Institute of Diabetes& Digestive & Kidney Diseases: Alternative Therapies for Diabetes |
http://www.niddk.nih.gov/health/ |
Clearinghouse (albeit abridged) for some alternative treatments for diabetes |
|
Supplement Watch |
A self-funded, privately held corporation consisting of scientists, physiologists, nutritionists, and other health professionals dedicated to educating the public about the pros and cons of dietary supplementation. |
|
|
Rick Mendosa's Diabetes Directory |
An extensive and unbiased private site published by a freelance journalist and consultant specializing in (and afflicted with) diabetes |
|
|
US FDA Office of Nutritional Products, Labeling, and Dietary Supplements |
FDA office responsible for developing policy and regulations for dietary supplements, medical foods, and related areas, as well as for their scientific evaluation |
|
|
WebMD Health |
WebMD provides comprehensive health information and tools for managing health care. For health care professionals and their patients |
|
Potential Cardioprotective Effects of Insulin Therapy
· There is a well known epidemiologic association between endogenous hyperinsulinemia and the incidence of coronary heart disease in non-diabetic individuals (14)
· Despite concerns that hyperinsulinemia may be related to increased cardiovascular risk, evidence from the Diabetes Mellitus Insulin Glucose Infusion in Acute Myocardial Infarction (DIGAMI) study demonstrated that exogenous insulin had a cardioprotective effect following acute myocardial infarction (MI); in this study, long term intensive insulin therapy had substantial cardiovascular benefits, resulting in a significant relative risk reduction in mortality by 28% versus conventional therapy at an average of 3.4-year follow-up (Figure 2a) (15; 16)
· This benefit was even more pronounced in patients who were insulin-naïve prior to MI, where the relative risk reduction for mortality was 51% with intensive insulin therapy (Figure 2b) (16)
· These results indicate that, in patients with Type 2 diabetes, particularly those who have a high cardiovascular risk, insulin therapy may be very relevant and needs to be seriously considered. Further long term cardiovascular outcome studies are underway to confirm these findings
|
|
|
Figure 2. Cardiovascular Benefit Of Intensive Insulin Therapy Post-MI In The DIGAMI Study. Patients (n=620) who had experienced MI were randomized to either intensive insulin therapy (48-hour continuous infusion followed by four-times daily insulin injections for up to 5 years) or conventional therapy (according to physician judgement). Panel A shows the total population; panel B shows insulin-naïve patients (n=272) [Taken from Malmberg et al, BMJ 314:1512–1515, 1997] Permission requested. |
· Furthermore, the UKPDS provided no evidence of an increased incidence of atherosclerotic events in people with Type 2 diabetes who were treated with insulin versus those who did not receive insulin (17)
· Over the years, multiple barriers to insulin therapy, such as fear of needles, limitations of insulin formulations, complexity of regimens and misconceptions relating to the significance of insulin with respect to complications, have challenged the acceptance and the effective use of insulin replacement therapy in Type 2 diabetes (13)
· Despite evidence that insulin therapy may be most beneficial in patients with Type 2 diabetes as outlined above, physicians also have numerous barriers to insulin therapy including (Table 1):
– Exaggerated concern that insulin therapy may cause significant weight gain in patients with Type 2 diabetes
– Valid concerns that the risk of hypoglycemia will be increased, which is partly related to the limitations of traditional insulin preparations
– Misconceptions from the epidemiologic non-interventional studies about endogenous insulin (not exogenous insulin therapy) as a factor that may increase cardiovascular risk
– Limited resources and time constraints to instruct patients with Type 2 diabetes how to inject and adjust insulin therapy
– Complex insulin strategies starting with multiple insulin injections
– Skepticism that patients will not accept insulin and will not follow titration algorithms to achieve stringent glycemic targets
|
Table 1. Barriers To Insulin Therapy In Type 2 Diabetes |
|
Patient Resistance |
|
|
Physician Resistance |
|
|
Physician and Patient Resistance |
|
· These barriers are discussed in detail below
· Initiation of insulin therapy is, typically, associated with weight gain, which correlates with increasingly better glycemic control
· This increase in weight in the range 2–5 kg may be partly explained by the fact that patients often maintain their previous inappropriate dietary regimens but no longer experience caloric loss from glycosuria, as a result of improved glycemic control (18)
· However, this weight gain has been shown to be insufficient to increase cardiovascular (CV) risk factors (e.g. blood pressure, lipid patterns) in obese patients with Type 2 diabetes; indeed, some risk factors were actually improved with insulin therapy (19-22)
· In addition, the weight gain associated with insulin therapy can be minimized by reducing caloric intake and increasing physical activity as well as combining with metformin, as will be discussed below
· Hypoglycemia is the single most important limiting factor to increasing and adjusting insulin doses to attain satisfactory glycemic control (13)
· Hypoglycemia risk is dependent on a number of factors, including age, weight, duration of diabetes, severity of insulin resistance, consistency of caloric intake, history of recurrent hypoglycemia, alcohol consumption, physical activity, degree of diabetic neuropathy and awareness of hypoglycemia
· It is clear that there is an extremely strong correlation between the level of glycemic control achieved in Type 1 diabetes and the resulting level of severe hypoglycemia, as evidenced by data from the Diabetes Control and Complications Trial (DCCT; Figure 3) that demonstrated a three fold increase in hypoglycemia with tight control (23)
· The UKPDS showed that, in patients with Type 2 diabetes, most incidents of hypoglycemia were mild in severity; severe hypoglycemia occurred in only 2–3% of patients each year (17; 24)
· However, it should be remembered that in the majority of studies in Type 2 diabetes, patients do not reach near-normoglycemia; therefore, it is entirely possible that hypoglycemia, severe or otherwise, would become more common as more patients are aggressively managed following recommended guidelines to achieve better glycemic control treating to target A1c levels (13)
|
|
|
Figure 3. Correlation Between Tighter Glucose Control (Lower Hemoglobin A1c) And Severe Hypoglycemia. [Taken from Am J Med 90:450–459, 1991] Permission requested. |
Limitations
of Traditional Human Insulin Formulations
· A comparison of the kinetics of the insulin preparations available for clinical use is described in Table 2 (13); this includes traditional formulations and new analogs of human insulin
· Regular human insulin is less than ideal for post-prandial glucose control:
– The slow onset of action necessitates administration 20–40 minutes before meals, putting patients at risk of pre-meal hypoglycemia if the meal is delayed
– The duration of action extends beyond the duration of endogenous insulin activity, therefore, the risk of hypoglycemia is increased; this encourages between-meal snacking, which promotes weight gain in Type 2 diabetes
· Insulin lispro and aspart are two short-acting insulin analogs with increased rate of dissociation into insulin monomers resulting in rapid absorption profiles that allow for more physiologic replacement of mealtime insulin requirements
· These two analogs developed for post-prandial blood glucose control,match normal mealtime insulin secretion patterns far more closely than does exogenous regular human insulin, with a more rapid onset of action, an earlier peak of activity and a shorter duration of action (25; 26)
· Clinical experience has shown that these properties result in improved post-prandial glucose control (26-32). Insulin glulisine is another fast-acting analog in clinical development that appears to have similar pharmacokinetic and pharmacodynamic profiles as insulin lispro and aspart
· However, their rapid waning of activity places greater dependence on sufficient interprandial basal insulin to maintain blood glucose control
· With increasing use of lispro and aspart, there is a growing need for protracted-acting insulins that afford adequate and reliable glycemic control in the post-absorptive state
|
Table 2. Pharmacokinetic Profiles Of Human Insulin Formulations And Analogs Currently Available For Clinical Use. Values are guidelines only, since time–action profiles will be subject to some degree of variability between and within individuals. [Taken from Rosenstock and Wyne, Textbook of Type 2 Diabetes. Goldstein B, Müller-Wieland D, Eds. London, Martin Dunitz, 2003, pages 131-154] |
|||
|
|
Onset of Action (h) |
Peak (h) |
Duration of Action (h) |
|
Insulin formulations |
|||
|
Regular |
0.5–1 |
2–4 |
6–10 |
|
NPH insulin |
2–4 |
4–8 |
12–16 |
|
Lente® |
2–4 |
4–8 |
12–16 |
|
Ultralente® |
4–6 |
Unpredictable |
18–20 |
|
Insulin analogs |
|||
|
Lispro (Humalog®) |
5–15 min |
1 |
4–5 |
|
Aspart (Novolog®) |
5–15 min |
1 |
4–5 |
|
Glargine (Lantus®) |
1–2 |
Flat |
~24 |
· Of the insulin formulations used traditionally for basal insulin replacement, NPH insulin and lente are intermediate-acting, with durations of action considerably less than 24 hours, and pronounced peaks in activity within a few hours of administration
· Ultralente, although longer acting, is subject to considerable variation in dosing, with erratic peaks in activity
· These limitations leave patients at increased risk of unpredictable hypoglycemia
· Insulin glargine is the first clinically available long-acting analog of human insulin (33)
· Insulin glargine (LANTUS) results from amino acid changes that shift the iso-electric point of the insulin molecule towards neutral, making it soluble at slightly acidic pH and less soluble at the physiologic pH of the subcutaneous tissue leading to microprecipitation which forms the basis for its protracted duration of action
· In contrast to the traditional intermediate- and long-acting insulin preparations described above, insulin glargine has a 24-hour flat time–action profile with no pronounced peaks
· This profile resembles that of continuous subcutaneous insulin infusion (CSII), which is considered to be the ‘gold standard’ of basal insulin replacement therapy (Figure 4) (34-36)
|
|
|
Figure 4. Plasma Glucose Concentrations after Subcutaneous Injection. Results of a glycemic clamp study. [Taken from Lepore et al, Diabetes 49:2142–2148, 2000] Permission requested |
· Furthermore, the duration of action of insulin glargine is more prolonged than those of ultralente and NPH insulin, at approximately 24 hours (Figure 5)
· Consistent with its slow absorption rate and duration of activity, insulin glargine allows for a once daily administration with no peak of action and less variability than NPH
· Clinical experience with insulin glargine has demonstrated that this pharmacokinetic and pharmacodynamic profile translates into a reduced risk of nocturnal hypoglycemia, without compromising glycemic control, particularly in Type 2 diabetes (37-43)
· An additional benefit for this analog is the consistency of absorption irrespective of injection site (arm, leg or abdomen) (44)
|
|
|
Figure 5. Time–Action Profile Of Insulin Glargine Following Subcutaneous Injection. Glycemic clamp study. [Taken from Lepore et al, Diabetes 49:2142–2148, 2000] Permission requested. |
· Insulin detemir is another long-acting insulin analog in final stages of clinical development and testing that is expected to be commercially available soon (45)
· The retardation principle for this analog is fatty acylation of the insulin molecule, which promotes albumin binding (46)
· Glycemic clamp studies have shown that, like insulin glargine, the profile of detemir is flatter than that of NPH insulin (47)
· However, the duration of action has been shown to be less than 24 hours at therapeutic doses, suggesting that this analog may not be suitable for once-daily dosing and will deliver its best effect with a twice-daily regimen (47)
· Administration of insulin by injection is inherently a barrier to use in patients with Type 2 diabetes, who are likely to have been taking only oral agents for several years before insulin therapy is initiated
· Developments in delivery devices such as pens that reduce the discomfort of injection and simplify the administration of insulin have addressed this issue (48; 49)
· Inhaled insulin is another delivery option that would remove the skin barrier of injection fear completely
· There is already proof of concept reported for the use of inhaled insulins for post-prandial blood glucose control (50; 51)
· However, this form of insulin delivery is still in the advanced clinical developmental stages, and longer studies assessing the impact of inhaled insulin on a number of therapeutic scenarios with particular attention to safety and efficacy are awaited with great interest
Misconception of Increased Cardiovascular Risk
· The association between insulin resistance (with the resulting endogenous hyperinsulinemia), central obesity, hypertension and dyslipidemia, all of which are known CV risk factors (14; 52), has raised some concern among physicians that insulin therapy may increase CV risk
· However, in fact the opposite is probably true: insulin may have beneficial cardioprotective effects in patients with Type 2 diabetes (see section 2.3)
Physiologic Insulin Replacement – The Basal/Bolus Concept
· Insulin replacement therapy should ideally mimic the physiologic insulin secretion patterns that are observed in response to 24-hour fasting post-absorptive and postprandial glucose profiles (Table 3)
|
Table 3. The Basal–Bolus Concept – Properties of the Basal and Bolus Insulins. [Adapted from Rosenstock and Wyne, Textbook of Type 2 Diabetes. Goldstein B, Müller-Wieland D, Eds. London, Martin Dunitz, 2003, pages 131-154] |
|
|
Insulin component |
Properties |
|
Basal |
· Near-constant insulin level throughout the day · Suppresses hepatic glucose production and lipolysis overnight and during prolonged periods between meals · Covers ~50% of daily insulin needs |
|
Bolus |
· Immediate rise and sharp peak at 1 hour · Limits post-meal hyperglycemia · Cover ~10–20% of total daily insulin requirement at each meal |
· The basal–bolus concept aims to mimic as closely as possible this complex physiologic daily pattern of insulin secretion in healthy individuals (Figure 6) (53)
|
|
|
Figure 6. Normal Physiologic Insulin Secretion. [Taken from Lee and Zinman, Diabetes Reviews Vol 6; 73–88, 1998] Permission requested. |
· The role of the basal insulin in basal–bolus regimens is to suppress hepatic glucose production and lipolysis in the post-absorptive state between meals and overnight
· The role of the bolus insulin is to limit hyperglycemia immediately after meals
· The basal–bolus concept is used routinely in Type 1 diabetes, but is also applicable to Type 2 diabetes, since both prandial and interprandial as well as fasting blood glucose levels are elevated in Type 2 diabetes (Figure 7)
|
|
|
Figure 7. Prandial And Interprandial Blood Glucose In Type 2 Diabetes Versus Healthy Individuals. [Taken from Riddle, Diabetes Care 16n:676–686, 1990] Permission requested. |
Earlier Insulin Replacement…When?
· The UKPDS Glucose Study 2 (UKPDS 57) provided strong evidence that the immediate addition of a basal insulin regimen when sulfonylurea monotherapy is unable to maintain fasting plasma glucose (FPG) <108 mg/dL substantially improves glycemic control, without significantly increasing the risk of hypoglycemia (24)
· In this long-term study, approximately 50% of patients receiving insulin added to a sulfonylurea achieved a median A1c level of 6.7% at 6 years, which is an important translational message to physicians as an effective management strategy to sustain glycemic targets in clinical practice
· Most importantly, in keeping with the results of the UKPDS Glucose Study 2, the concept of insulin as expected therapy in the management of Type 2 diabetes should be introduced at the time of diagnosis, so the patient knows and understands better the role of insulin and will eventually accept insulin therapy more openly when overtime glycemic control can not be adequately maintained by oral agents
· The mental barrier and misconceptions that insulin therapy is a sign of failure, or indicative that finally the Type 2 diabetes “has worsened or has become serious”, need to be permanently dispelled
· At the present time, a practical approach to encourage earlier insulin initiation is that insulin therapy should be considered if the A1c levels remain >7%, despite maximized oral agent therapy and lifestyle intervention
· Maximizing oral therapy could be one, two or three oral agents according to the patients’ tolerance and side effect profile
· Conventionally, maximized oral agent therapy is a combination of three oral agents; however, the effectiveness in reaching and sustaining A1c targets is relatively limited as shown below (Table 4)
|
Table 4. Effectiveness of Three Oral Agents for Reaching A1c Targets |
||
|
Class of oral agent |
SU + Metformin + Troglitazone (54) |
Glyburide + Metformin + Rosiglitazone (55) |
|
Baseline A1c
|
9.7%
|
8.1% |
|
A1c reduction from baseline |
-1.4%
|
-0.9%
|
|
Reached target A1c <7% |
14% |
42% |
|
SU=sulfonylurea |
||
Limitations of Conventional Pre-Mixed Formulations
· The use of ‘pre-mixed’ insulin preparations, such as 70/30 or 75/25, using NPH insulin plus regular, lispro or aspart do not provide enough flexibility and are seldom effective for reaching glycemic targets
· Additional limitations of pre-mixed insulins are listed in Table 5:
|
Table 5. Practical Limitations To Pre-Mixed Insulins |
|
Practical Limitations to pre-mixed insulins |
|
· The twice-daily ‘split-mixed’ regimen consists of a morning and evening mix of NPH insulin and a short-acting insulin
· The short-acting insulin (regular, lispro or aspart) in the morning covers the time between breakfast and lunch, whereas the intermediate-acting NPH insulin peaks around noon and covers the afternoon until dinnertime
· The second ‘split-mixed’ dose administered at the evening meal, provides insulin coverage between the dinner and bedtime interval (short-acting insulin component) and also coverage overnight (NPH insulin component); this overnight NPH insulin dose often peaks during the late night
· Furthermore, these two insulins have overlapping peaks, which can occur at any time, resulting in frequent periods of inappropriate hyperinsulinemia and a propensity to hypoglycemia, which can lead to frequent snacking
· The 70/30 pre-mixed NPH/regular insulin, 70/30 pre-mixed NPH/aspart insulin or 75/25 pre-mixed NPL/lispro have been traditionally used. However, the fundamental problem with these preparations is the rigidity of the supplementation and lack of flexibility for specific insulin adjustments because the two insulins are adjusted at the same time based on self-blood glucose monitoring. These pre-mixed insulins often result in periods of excessive hyperinsulinemia with increased risk of hypoglycemia which is a major limitation for reaching glycemic targets (Table 6)
|
Table 6. Comparison of Triple Oral Therapy with Insulin Premix plus Metformin (56) |
||
|
|
SU + met + TZD (Triple Oral Therapy) |
Metformin + Insulin 70/30* |
|
Baseline A1c A1c reduction from baseline Reached target A1c <7% Hypoglycemia (<50 mg/dL) |
9.6% -1.7% 31% 48% |
9.7% -1.9% 32% 67% |
|
*NPH insulin + regular insulin; SU=sulfonylurea; met=metformin; TZD=thiazolidinedione |
||
Initiating Insulin Therapy with Basal Insulin
Rationale for Starting with Basal Insulin as Add-on to Oral Therapy
· Basal insulin replacement therapy covers approximately 50% of the daily insulin needs and should ideally mimic the physiologic insulin secretion patterns to control the 24-hour post absorptive glucose profiles by suppressing hepatic glucose production overnight, during fasting and during prolonged periods between meals
· A single evening or bedtime injection of basal insulin with continued use of one or more oral agents has been shown to lower fasting hyperglycemia, with a beneficial carryover effect on glycemic levels later in the day; this results in significant improvements in A1c levels
· One explanation for this outcome is that basal insulin can improve overnight and fasting glucose control enough to decrease glucotoxicity, allowing oral agents to have their full effect on modulating and increasing insulin secretion for mealtime control
· The cumulative impact is improved glycemic control with a reduced need for exogenous insulin as endogenous insulin secretion increases as a result of preserved β cell function and possible β cell rest (13; 57)
Practical Advantages of Early Basal Insulin Add-on to Oral Therapy
· This simple approach overcomes the complexity of insulin regimens in Type 2 diabetes, and is perhaps the most practical and most acceptable way to initiate insulin therapy, by just adding an evening basal insulin replacement in patients who are no longer responding to oral agents
· There are numerous benefits in adding a basal insulin replacement while maintaining the use of oral antidiabetic agents:
– Only one daily insulin injection may be required, thus keeping the treatment regimen simple, and the mixing of different insulin preparations is not needed
– Easy titration in a slow and, therefore, safer way but on a consistent basis according to fasting blood glucose targets
– Administration with an insulin pen facilitates and enhances patient acceptance and adherence to insulin therapy
– Eventually, basal insulin used in combination with oral antidiabetic agents will require a lower total insulin dose because of the synergy or complementary effects of individual oral therapies
· These advantages result in a greater improvement in glycemic control compared with the use of oral antidiabetic agents alone with only limited weight gain
· The ability to add a single basal insulin dose to the therapeutic regimen of Type 2 diabetes patients acts as a simple and effective ‘bridge strategy’ that will be more acceptable to many patients and will facilitate and enable them to overcome their resistance to insulin therapy
· Eventually, if the A1c levels remain >7% despite the use of evening basal insulin in combination with oral therapy , and as long as the fasting glucose levels are normal, then further benefit can easily be obtained by progressively adding pre-meal lispro or aspart insulin to the main meals to control postprandial hyperglycemia
· There is much clinical evidence to support the safety and efficacy of the addition of basal insulin to oral agents, which is summarized in the next section
Clinical Experience of Initiating Insulin Therapy with Basal Insulin
· The synergistic effects of basal insulin added in combination with insulin secretagogues and insulin sensitizers have been demonstrated in a number of studies adding once-daily ultralente to sulfonlyureas, bedtime NPH insulin to glipizide or glyburide, dinnertime 70/30 mix to glimepiride, or bedtime NPH insulin to various oral agents including metformin
Sulfonlyurea
Plus Basal Insulins
|
Table 7. Clinical Experience with Sulfonylurea Plus Ultralente |
|
|
Study (reference) |
Key findings |
|
Oxford Study: Randomized, crossover study, 15 patients, 8-week treatment, ultralente vs. ultralente + sulfonylurea (58) |
|
|
UKPDS Glucose Study 2: Conventional (diet) vs. sulfonylurea + ultralente vs. ultralente alone in 826 newly diagnosed patients (24) |
|
· Conclusion: Addition of long-acting ultralente to sulfonylurea effectively restores normoglycemia, particularly when used early in disease progression, without increasing hypoglycemia or weight gain
|
Table 8. Clinical Experience with Sulfonylurea plus Bedtime NPH insulin |
|
|
Study (reference) |
Key findings |
|
San Antonio Study: Double-blind study of sulfonylurea + NPH vs. sulfonylurea alone vs. NPH alone in 30 patients with secondary sulfonylurea failure (59) |
· Markedly reduced fasting plasma glucose, mean 24-hour glucose, A1c and basal hepatic glucose production with NPH + sulfonylurea vs. sulfonylurea or NPH insulin alone · Further reduction of A1c with progressive dose titration of insulin in NPH + sulfonylurea combination · Glycemic control maintained for up to 1 year with NPH + sulfonylurea |
· Conclusion: Addition of bedtime NPH insulin to sulfonylurea effectively improves glycemic control and overcomes secondary sulfonylurea failure
Different Oral Agents Plus NPH Insulin
|
Table 9: Clinical Experience with Metformin +/- Sulfonylurea plus Bedtime NPH |
|
|
Study (reference) |
Key findings |
|
FINFAT Study: One year study in 96 patients inadequately controlled on sulfonylurea randomized to: bedtime NPH insulin + sulfonylurea vs. bedtime NPH insulin + metformin vs. bedtime NPH insulin + sulfonylurea and metformin vs. bedtime NPH insulin + morning NPH insulin (60) |
|
· Conclusion: Proof of concept of the effectiveness and value of combination oral agents and basal insulin therapy. Bedtime NPH insulin + metformin afforded greater improvement in glycemic control with less weight gain. However, the combination of sulfonylurea and bedtime NPH insulin was slightly less effective but with less hypoglycemia, probably owing to lower insulin doses
Oral agents Plus Insulin Glargine (LANTUS)
· Insulin glargine, a long-acting human insulin analog with a flat profile of action, is being increasingly used to treat Type 2 diabetes patients
· Glargine may be especially suited as a supplemental basal insulin added to oral therapy owing to a lower risk of nocturnal hypoglycemia compared with NPH insulin (42; 43)
· The “One Pill-One Shot” study compared combinations of the oral agent glimepiride with bedtime NPH insulin versus glimepiride with bedtime or morning insulin glargine in patients with Type 2 diabetes previously treated with one or two oral agents(61)
· All three groups were on glimepiride but the risk of nocturnal hypoglycemia was lower in the groups with morning insulin glargine (17%) and bedtime insulin glargine (23%) than with bedtime NPH insulin (38%)
· A1c levels improved by –1.24% with morning insulin glargine, –0.96% with bedtime insulin glargine and –0.84% with NPH insulin over the 6 month study but it is conceivable that the results could have been more effective if metformin or an insulin sensitizer had been used in combination with the sulfonylurea
· The ‘Treat-to-Target’ study, which compared basal insulin replacement therapy with either bedtime insulin glargine or NPH insulin added to oral combination therapy also in insulin-naïve patients, revealed the efficacy of these insulins administered using a structured titration algorithm for actively adjusting insulin doses to achieve fasting plasma glucose (FPG) levels <100 mg/dL and reaching A1c levels <7% (62)
· A1c values decreased from 8.6% at baseline to 6.9% by the end of the 6 month study with both insulins, successfully achieving A1c targets in almost 60% of patients
· However, the patients receiving insulin glargine experienced a 44–48% risk reduction in nocturnal hypoglycemia compared with those receiving NPH insulin (Figure 8)
· Furthermore, significantly more patients in the insulin glargine group (33%) reached target A1c ≤7%, without a single episode of documented nocturnal hypoglycemia, compared with those in the NPH insulin group
|
|
|
Figure 8. Comparison of hypoglycemia profiles with NPH insulin versus insulin glargine in combination with oral agents, in patients with Type 2 diabetes. [Adapted from Riddle et al, Diabetes Care 26:3080-3086, 2003] Permission requested. |
· Type 2 diabetes patients treated to glycemic target are always at increased risk of hypoglycemia and thus any therapeutic tool that has less risk of hypoglycemia is a welcome pharmacologic addition
Initiating Insulin Therapy with Bolus Insulin
Rationale for Initiation of Bolus Insulin as Add-On to Oral Therapy
· The rationale for initiation of insulin therapy with mealtime insulin supplementation is to improve postprandial hyperglycemic peaks while continuing sensitizing agents such as metformin or a glitazone with or without sulfonylureas
Clinical Experience Of Initiating Insulin Therapy With Bolus Insulin
· Initiating a bolus insulin for patients with Type 2 diabetes for whom oral therapy is no longer able to maintain normoglycemia has been shown to be effective with three injections of insulin lispro added to sulfonylurea (63)
· This study showed that the addition of preprandial insulin lispro resulted in lower A1c (7.6%) compared with the addition of bedtime NPH insulin or the addition of metformin in a 12-week time period
· The morning FBG was lowest in the bedtime NPH insulin group; however, postprandial glucose was lowest in the insulin lispro + sulfonylurea group
· The feasibility of using inhaled insulin to initiate bolus insulin therapy has been demonstrated in Type 2 diabetes patients who did not maintain A1c <8.0% on combination oral therapy (64)
· This large study demonstrated that patients receiving combination oral therapy together with inhaled insulin experienced the greatest decrease in A1c (-1.9%) compared with those receiving inhaled insulin alone (-1.4%) or those patients who continued therapy with oral agents alone (-0.2%) (64)
· Further long-term studies to examine the pulmonary safety and sustained efficacy of inhaled insulin are required before FDA approval and it is made commercially available to patients
Practicalities Of Initiating Insulin With Bolus Insulin
· Initiating insulin as bolus therapy in patients with Type 2 diabetes is not optimal owing to the need for multiple premeal injections, which means the regimen is more complex and, therefore, less attractive to Type 2 diabetes patients compared with introducing a single basal insulin injection
· The potential availability of inhaled insulin some time in the future may enable bolus insulin supplementation in Type 2 diabetes patients to become a reality, as this type of administration is likely to be more acceptable to patients
Insulin Replacement Therapy – Practical Guidelines
· In light of the increasing tendency to initiate insulin therapy much earlier in the natural clinical progression of Type 2 diabetes, there is a need for practical guidelines to enable general physicians to administer insulin effectively and safely
Practical Guidelines For Starting Basal Insulin
· As discussed above, probably the most practical way to initiate insulin therapy in Type 2 diabetes is the addition of an evening basal insulin supplement to ongoing oral therapy
· The efficacy, safety and ease of this strategy make it especially accommodating to the needs of time-pressed clinicians in general practice, who manage the vast majority of patients with Type 2 diabetes
· Early initiation of basal insulin, as a supplement to ongoing oral therapy, is a simple and effective strategy, which may help patients with Type 2 diabetes overcome their resistance to insulin therapy
· This approach also alleviates key patient concerns about insulin therapy, dispelling the myths that it is too complex and poses a dangerously high risk of hypoglycemia
· Insulin glargine (LANTUS), by virtue of its long duration, flat profile of action and reduced risk of nocturnal hypoglycemia, has the potential to enable a simple initiation of insulin therapy
· Insulin glargine was initially studied using a bedtime once-daily administration but, as a result of its pharmacokinetic profile, it could theoretically be given at any time of the day (although at the same time every day) (65)
· Effective titration of basal insulin while continuing oral therapy, can be achieved on the basis of self-monitored FBG levels
· The dose increase would depend on the FBG according to a titration algorithm that is continuously followed by the patient to sustain glycemic targets
· If levels are elevated above a pre-defined target FBG level, usually <100 mg/dL, as an average from two or three consecutive days, the insulin dose is then increased, provided nocturnal hypoglycemia has not occurred and/or the FBG is not <72 mg/dL
· This method of titration was employed in the Treat-to-Target study with a pre-defined FBG treatment target of 100 mg/dL (62); the starting dose of basal insulin was a once-daily evening dose of 10 IU and a weekly forced titration schedule as shown in Table 10
|
Table 10. Practical Guidelines for Starting Basal Insulin from the Treat To Target Study |
|
Practical guidelines for starting basal insulin from the Treat To Target Study |
|
|
|
|
|
|
FBG=fasting blood glucose |
· Alternatively, an even simpler titration algorithm that is well accepted and facilitates the ‘empowerment’ of the patients to maintain patient-driven basal insulin adjustments to reach and sustain FBG <100 mg/dL, as long as there are no symptoms or evidence of hypoglycemia, is as follows:
|
Table 11. Practical Guidelines for Starting Basal Insulin |
|
Practical guidelines for starting basal insulin |
|
|
|
|
|
FBG=fasting blood glucose |
· Oral agents can be of value at this point, depending on individual patient needs and side effect profile (Table 12)
|
Table 12. Basis to maintain Oral Agents in Combination with Insulin Therapy |
|
|
Oral agent |
Patient characteristics |
|
Metformin |
|
|
Thiazolidinedione |
|
|
Secretagogues |
|
· Metformin therapy can be continued in parallel with basal insulin therapy in order to control weight gain as long as renal impairment and congestive heart failure (CHF) are not present
· A glitazone may be used to reduce insulin resistance and potentially preserve beta cell function, as long as there is no evidence of substantial weight gain and fluid retention
· Secretagogues can be potentially useful to enhance endogenous insulin secretion, especially after glucotoxicity has been corrected and following ‘beta cell rest’ with exogenous insulin; the response to sulfonylureas or meglitinides may control postprandial hyperglycemia better
Practical Guidelines for Advancing to Basal–Bolus Therapy
· Over time, insulin regimens will need to be intensified in response to disease progression as determined by the lack of attainment or the loss of sustained glycemic targets
· As Type 2 diabetes progresses, and beta cell function declines, there will be an eventual need to intensify therapy as oral agents begin to lose the ability to control postprandial hyperglycemia
· When a target FPG of <100 mg/dL has been achieved and the A1c is still above 7.0%, further increments in the dose of basal insulin glargine may be employed; however, this approach could result in an increased risk of hypoglycemia
· For these patients, additional insulin, basal or prandial, can be incorporated into the treatment regimen as shown in Table 12, depending on the basal insulin type used
· Pre-mixed insulin preparations do not provide the necessary flexibility to achieve the recommended levels of glycemic control and should preferably not be considered when other insulin therapies are available
· The next step, therefore, is to introduce a preprandial fast-acting insulin analog at the main meal in the first instance, which should result in an improvement in postprandial glucose levels
· Multiple daily insulin regimens (MDI) using short-acting insulin analogs (lispro or aspart) at the main meal or at each meal, to closely match physiologic prandial insulin patterns, can further optimize the once-daily basal insulin glargine therapy, which provides a more predictable basal control than NPH insulin with less risk of nocturnal hypoglycemia
· The titration schedule should be based on self-monitored blood glucose levels according to some predetermined postprandial glucose levels; controversies still exist regarding the target 2-hour postprandial levels that are usually in the 140–180 mg/dL range
· By titrating basal and bolus insulin separately, maximum treatment flexibility is achieved
· Using this approach, multiple dose insulin (MDI) therapy is progressively introduced to the patient in order to ultimately mimic normal physiology (Table 13)
|
Table 13. Intensifying Basal Insulin Therapy with Bolus Insulin [Adapted from Rosenstock and Wyne, Textbook of Type 2 Diabetes. Goldstein B, Müller-Wieland D, Eds. London, Martin Dunitz, 2003, pages 131-154] |
|
Practical guidelines for advancing to basal/bolus insulin |
|
|
|
|
FBG=fasting blood glucose; SMBG=self-monitored blood glucose |
· This practical and simple strategy outlined above should enable Type 2 diabetes patients to reach and sustain defined glycemic targets, facilitating the translation of clinical research studies into clinical practice
The Future of Insulin Therapy in Type 2 Diabetes
· Insulin therapy in Type 2 diabetes has been viewed traditionally as a last resort used only when all other therapeutic options have been exhausted
· Physicians must, therefore, view insulin as a vital therapeutic tool for attaining and sustaining treatment goals, not as a sign of treatment failure
· A shift in the treatment paradigm for Type 2 diabetes is needed, towards the earlier use of insulin to preserve beta cell function to maintain long term near-normoglycemic control only limited mainly by hypoglycemia
· Hopefully, the simple strategies reviewed above will evolve to become the standard of care to benefit patients with Type 2 diabetes. The early use of supplemental basal insulin glargine to oral combination therapy to reach A1c targets and if needed, systematically adding a meal-time short-acting analogue lispro or aspart insulin to control postprandial hyperglycemia, has the potential to be widely translated to the primary care settings
· The use of effective insulin pens is increasing and will clearly improve the adherence and compliance of the patients to follow insulin replacement strategies
· In order to maintain good glycemic control in the long term, insulin doses should be continuously titrated to reach treatment glycemic targets, with special consideration always given to hypoglycemia
· With continuing improvements in insulins (both pharmacokinetic properties and delivery routes) and treatment regimens, the ultimate goal of reduced risk of long-term complications through tight glycemic control will be achievable
The National Society to Prevent Blindness has estimated that 4 to 6 million diabetics in the U.S. have diabetic retinopathy. In the Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR), a population-based study, the prevalence of diabetic retinopathy was evaluated in patients diagnosed with diabetes before and after age 30. In the younger group, in which all the patients received insulin therapy, retinopathy was present in 13% of patients whose diabetes had been present for less than 5 years, whereas over 90% of patients with diabetes for 10 to 15 years had retinopathy. In the older group, 40% of patients using insulin and 24% of those not receiving insulin had retinopathy after less than 5 years of diabetes. In older patients with diabetes for 15 or more years, 84% of patients on insulin and 53% of those not using insulin had retinopathy.(1)
The WESDR also showed that the rate of vision loss increased with the severity of retinopathy and with the duration of diabetes.(2) In patients diagnosed before age 30, 3% of those with diabetes for 15 to 19 years were blind, as were 12% of those with diabetes for 30 years or more. In all diabetics aged 65 to 74 years old, 14% of males and 20% of females were legally blind. Over a 10-year period, 9% of the younger-onset patients had doubling of the visual angle (e.g. a drop from 20/40 to 20/80 on Snellen acuity testing), compared with 32% of older patients on insulin and 21% of older patients not on insulin.(3) Another study found that diabetes increased the rate of legal blindness by a factor of 50 to 80.(4)
In caring for diabetic patients, therefore, health care providers must bear in mind the substantial risks of developing visual loss that these patients face. For affected patients, diabetes-related visual loss decreases the quality of life and interferes with the performance of daily activities. On a larger scale, it is estimated to cost the U.S. $500 million per year.(5) An understanding of the factors predisposing to the development and worsening of retinopathy can help practitioners delay or prevent its onset.
The principal factor related to the development or worsening of diabetic retinopathy is glucose control. The Diabetes Control and Complications Trial (DCCT), a randomized, controlled study of 1441 patients with type 1 diabetes found that an intensive glucose control regimen reduced the risk of developing retinopathy by 76%.(6) In patients with pre-existing retinopathy, intensive control slowed progression of the condition by 54%. An analysis of HbA1C levels revealed that a 10% decrease in HbA1C resulted in a 35% to 40% reduction in the risk of worsening of retinopathy.(1)
Patients with type 2 diabetes were evaluated in the United Kingdom Prospective Diabetes Study (UKPDS). The study found a 25% reduction in the risk of microvascular endpoints, including the need for diabetic retinal laser treatment, with intensive glucose control.(1) Like the DCCT, the UKPDS also showed that decreasing the HbA1C lowered the risk of microvascular complications substantially. Of note, during 10 years of followup in the UKPDS and 9 years in the DCCT, retinopathy could not be completely prevented, even in the intensive therapy groups.
Despite the well-established benefits of tight glucose control for achieving favorable outcomes with diabetic retinopathy, patients must be advised that an initial worsening of retinopathy can occur when a more intensive glycemic control regimen is implemented. The DCCT and other smaller trials have found that in both type 1 and 2 diabetics, these changes can occur over the first 3 to 12 months after the glucose levels are controlled. For many patients, the changes are not clinically significant, but for patients with mild to moderate levels of retinopathy at baseline, the worsening of the retinopathy potentially can result in a decline in visual acuity, requiring intervention with laser treatment. In the DCCT, 19% of patients with moderate nonproliferative retinopathy experienced this early worsening effect. The reasons for this phenomenon are not understood. Observers agree that although these risks are outweighed by the long-term benefits in preventing severe vision loss from retinopathy, patients with retinopathy at baseline should be observed closely in the initial period following implementation of tighter glucose control.(7)
Other risk factors for progression of diabetic retinopathy include hypertension, hypercholesterolemia, and pregnancy. Both the WESDR and the UKPDS have established relationships between hypertension and worsening of diabetic retinopathy.(2) Strict blood pressure control with either atenolol or captopril was found to reduce the need for laser treatment by 35% compared to less rigorous control. Elevated cholesterol and triglycerides are also associated with retinopathy progression. Finally, pregnancy is a significant risk factor for worsening; type 1 diabetics are twice as likely to progress to proliferative disease if they are pregnant. Smoking, although a known risk factor for cardiovascular disease, did not correlate with increased likelihood of retinopathy progression in the UKPDS and WESDR.
The American Academy of Ophthalmology has recommended screening for diabetic retinopathy 5 years after diagnosis in patients with type 1 diabetes, and at the time of diagnosis in patients with type 2 diabetes. Patients without retinopathy should undergo dilated fundus examination annually. If mild nonproliferative diabetic retinopathy (NPDR) is present, exams should be repeated every 9 months. Patients with moderate NPDR should be examined every 6 months. In severe NPDR, exams should be conducted every 3 months. Patients with proliferative diabetic retinopathy should be examined every 2 to 3 months. During pregnancy, patients should be examined every 3 months, since retinopathy can progress rapidly in this setting.(16)
Various mechanisms account for the features of diabetic retinopathy. Histopathologic analysis shows thickening of capillary basement membranes, microaneurysm formation, loss of pericytes, capillary acellularity, and neovascularization.(8) Microaneurysms, outpouchings of the capillary wall, serve as sites of fluid and lipid leakage, which can lead to the development of diabetic macular edema. Theories on the biochemistry of these end-organ changes include toxic effects from sorbitol accumulation, vascular damage by excessive glycosylation with crosslinking of basement membrane proteins, and activation of protein kinase C-ß2 by vascular endothelial growth factor (VEGF), leading to increased vascular permeability and endothelial cell proliferation. VEGF, produced by the retina in response to hypoxia, is believed to play a central role in the development of neovascularization.(9)
|
|
|
Figure 1a. Microaneurysms:focal outpouchings of the capillary wall. |
CLINICAL FEATURES: NONPROLIFERATIVE DIABETIC RETINOPATHY (NPDR)
Studies have found that retinopathy in both insulin-dependent and non-insulin-dependent diabetes occurs 3 to 5 years or more after the onset of diabetes. In the WESDR, the prevalence of at least minimal retinopathy was almost 100% after 20 years.(8) A more recent study has confirmed that at least 39% of young diabetics developed retinopathy within the first 10 years.(15) The earliest clinical sign of diabetic retinopathy is the microaneurysm, a red dot seen on ophthalmoscopy that varies from 15 to 60 microns in diameter.
|
|
|
Figure 1b. Microaneurysms and intraretinal hemorrhages in nonproliferative retinopathy. (UCSF Department of Ophthalmology) |
The lesions can be difficult to distinguish from intraretinal hemorrhages on examination, but with fluorescein angiography microaneurysms can be identified easily as punctate spots of hyperfluorescence. By contrast, hemorrhages block the background fluorescence and therefore appear dark.
|
|
|
Figure 2. Microaneurysms: hyperfluorescent dots in early phase of fluorescein angiogram (arrows). (San Francisco General Hospital, Dept. of Ophthalmology) |
|
|
|
Figure 3. Two minutes later, fluorescein leakage from the microaneurysms gives them a hazy appearance. (San Francisco General Hospital, Dept. of Ophthalmology) |
The severity of NPDR can be graded as mild, moderate, severe, or very severe. In mild disease, microaneurysms are present with hemorrhage or hard exudates (lipid transudates). In moderate NPDR, these findings are associated with cotton-wool spots (focal infarcts of the retinal nerve fiber layer or areas of axoplasmic stasis) or intraretinal microvascular abnormalities (vessels that may be either abnormally dilated and tortuous retinal vessels, or intraretinal neovascularization). The "4-2-1 rule" is used to diagnose severe NPDR: criteria are met if hemorrhages and microaneurysms are present in 4 quadrants, or venous beading is present in 2 quadrants, or moderate intraretinal microvascular abnormalities are present in 1 quadrant. In very severe NPDR, two of these features are present.
The correct evaluation and staging of NPDR is important as a means of assessing the risk of progression. In the ETDRS, eyes with very severe NPDR had a 60-fold increased risk of developing high-risk proliferative retinopathy after 1 year compared with eyes with mild NPDR.(1) For eyes with mild or moderate NPDR, early treatment with laser was not warranted, as the benefits in preventing vision loss did not outweigh the side effects. By contrast, in very severe NPDR, early laser treatment was often helpful.
|
|
|
Figure 4. Venous beading (arrows) in a case of proliferative diabetic retinopathy. (UCSF Department of Ophthalmology) |
Vision loss in nonproliferative diabetic retinopathy is most commonly caused by macular edema. Because of the increased vascular permeability and breakdown of the blood-retinal barrier, fluid and lipids leak into the retina and cause it to swell. This causes photoreceptor dysfunction, leading to vision loss when the center of the macula, the fovea, is affected. In the ETDRS, macular edema was characterized as "clinically significant" if any of the following were noted: retinal thickening within 500 microns of the fovea, hard exudates within 500 microns of the fovea if associated with adjacent retinal thickening, or an area of retinal thickening 1 disc diameter or larger if any part of it is located within 1 disc diameter of the fovea.(1)
|
|
|
Figure 5. Clinically significant macular edema with hard exudates in the fovea. Cotton-wool spots are present near the major retinal vessels (arrows). (UCSF Dept. of Ophthalmology) |
Capillary closure can also result in macular ischemia, another cause of vision loss in NPDR. This can be identified clinically as an enlargement of the normal foveal avascular zone on fluorescein angiography.
|
|
|
Figure 6. Capillary dropout around the fovea (white arrow) and in the temporal macula (black arrow). (San Francisco General Hospital, Dept. of Ophthalmology) |
CLINICAL FEATURES: PROLIFERATIVE DIABETIC RETINOPATHY (PDR)
In proliferative diabetic retinopathy, many of the changes seen in NPDR are present in addition to neovascularization that extends along the surface of the retina or into the vitreous cavity. These vessels are in loops that may form a network of radiating spokes or may appear disorganized. In many cases the vessels are first noted on the surface of the optic disc, although they can be easily missed due to their fine calibur.(10) Close inspection often reveals that these new vessels cross over both the normal arteries and the normal veins of the retina, a sign of their unregulated growth.
|
|
|
Figure 7. Active neovascularization in PDR. Fibrovascular proliferation overlies the optic disc (white arrow). Loops of new vessels are especially prominent superior to the disc and extending into the macula, where leakage of fluid has led to deposition of a ring of hard exudate around the neovascular net (black arrow). (UCSF Department of Ophthalmology) |
New vessels can also appear on the iris, a condition known as rubeosis iridis. When this occurs, careful inspection of the anterior chamber angle is essential, as growth of neovascularization in this location can obstruct aqueous fluid outflow and cause neovascular glaucoma.
|
|
|
Figure 8. Rubeosis iridis in a case of PDR. Abnormal new vessels are growing along the surface of the iris (arrows). (UCSF Dept. of Ophthalmology) |
Neovascularization can remain relatively stable or it can grow rapidly; progression can be noted ophthalmoscopically over a period of weeks. The vessels often develop an associated white, fibrous tissue component that can increase in size as the vessels regress. The resulting fibrovascular membrane may then develop new vessels at its edges. This cycle of growth and fibrous transformation of diabetic neovascularization is typical.(7) The proliferation occurs on the anterior surface of the retina, and the vessels extend along the posterior surface of the vitreous body. Fibrous proliferation takes place on the posterior vitreous surface; when the vitreous detaches, the vessels can be pulled forward and the thickened posterior vitreous surface can be seen ophthalmoscopically, highlighted areas of fibrovascular proliferation.
The severity of PDR can be classified as to the presence or absence of high-risk characteristics. As determined in the Diabetic Retinopathy Study, eyes are classified as high-risk if they have 3 of the following 4 characteristics: the presence of any neovascularization; neovascularization on or within 1 disc diameter of the optic disc; a moderate to severe amount of neovascularization (greater than 1/3 disc area neovascularization of the disc, or greater than 1/2 disc area if elsewhere), or vitreous hemorrhage.(7)
Vision loss in proliferative diabetic retinopathy results from three main causes. First, vitreous hemorrhage occurs because the neovascular tissue is subject to vitreous traction. Coughing or vomiting may also trigger a hemorrhage. Hemorrhage may remain in the preretinal space between the retina and the posterior vitreous surface, in which case it may not cause much vision loss if located away from the macula. In other cases, though, hemorrhage can spread throughout the entire vitreous cavity, causing a diffuse opacification of the visual media with marked vision loss.
|
|
|
Figure 9. Preretinal hemorrhage: blood trapped between the retina and the vitreous in a case of incomplete vitreous detachment. Visual acuity is unaffected. (UCSF Department of Ophthalmology) |
|
|
|
|
|
Figure 10. Above: moderate vitreous hemorrhage; vision = 20/150. Below: 1 year later after spontaneous clearing of the hemorrhage; vision = 20/30. (San Francisco General Hospital, Dept. of Opthalmology) |
|
|
|
Figure 11. Dense vitreous hemorrhage almost completely obscuring the view of the fundus. (San Francisco General Hospital, Dept. of Ophthalmology) |
Another cause of severe vision loss in PDR is retinal detachment. As the fibrovascular membranes and vitreous contract, their attachments to the retina can cause focal elevations of the retina, resulting in a traction retinal detachment. In other cases the retinal vessels can be avulsed or retinal holes may be created by this traction, leading to a combined traction-rhegmatogenous retinal detachment.
|
|
|
Figure 12. Marked fibrosis with traction exerted on the retina outside the macula (arrows). The macula does not appear to be elevated. (UCSF Dept. of Ophthalmology) |
|
|
|
Figure 13. Traction retinal detachment outside the macula. Note elevation of retinal vessel out of the plane of focus (white arrow). Scatter photocoagulation scars are seen peripherally (black arrow). (UCSF Dept. of Ophthalmology) |
Finally, patients with PDR may have macular nonperfusion or coexisting diabetic macular edema that causes vision loss through photoreceptor dysfunction.
TREATMENT: LASER PHOTOCOAGULATION FOR NPDR
Tight glucose and blood pressure control are critical systemic factors in controlling the progression of diabetic retinopathy. Ocular complications of diabetes are addressed directly through treatment with laser photocoagulation or surgery.
Diabetic macular edema is believed to result from fluid and lipid transudation from microaneurysms and telangiectatic capillaries. Focal laser photocoagulation is used to heat and close the microaneurysms, causing them to stop leaking. Macular edema often improves following this form of treatment. Some clinicians apply laser burns in a grid pattern overlying areas of retinal edema without directing treatment to specific microaneurysms; this method can also be effective in reducing retinal thickening. The mechanism by which grid laser treatment achieves these results is not known.(10)
The ETDRS found that the risk of moderate visual loss in eyes with diabetic macular edema was reduced by 50% by photocoagulation. At 3 years, 24% of untreated eyes experienced a 3-line decrease in vision compared with 12% of treated eyes.(1) Only a very small percentage of eyes improved with treatment, highlighting the fact that the goal of treatment is not to improve vision, but rather to stabilize it and prevent worsening. Eyes meeting the criteria for clinically significant macular edema in which the edema was closest to the center were most likely to benefit from treatment. Side effects of laser treatment include scotomas, noticeable immediately after the procedure; late enlargement of laser scars can also occur, causing delayed visual loss. Inadvertent photocoagulation of the fovea is a risk of the procedure.(1) Since the amount of energy used is minimal, the treatment is performed under topical anesthesia.
TREATMENT: LASER PHOTOCOAGULATION FOR PDR
|
|
|
Figure 14. Focal laser scars in the macula following treatment for macular edema (arrow). Edema has resolved. (San Francisco General Hospital, Dept. of Ophthalmology) |
Scatter laser photocoagulation, also known as panretinal photocoagulation (PRP), is an important treatment modality for PDR and severe NPDR. Laser spots are placed from outside the major vascular arcades to the equator of the eye, with burns spaced approximately 1/2 to 1 burn width apart. Although the treatment destroys normal retina, the central vision is unaffected since all spots are placed outside the macula. The theory underlying this treatment is that photocoagulation of the ischemic peripheral retina decreases the elaboration of vasoproliferative factors contributing to PDR. Indeed, VEGF levels in the vitreous are increased in eyes with neovascularization, and they are lower after scatter photocoagulation.(11) Other factors such as insulin-like growth factor-1 are similarly elevated in the vitreous of eyes with PDR.(12)
Side effects of scatter photocoagulation include decreased night vision and dark adaptation, and visual field loss. The procedure can be painful, so treatment may be divided into several sessions, and either topical or retrobulbar anesthesia may be used.
|
|
|
Figure 15. Scatter photocoagulation scars in an eye with active PDR. Note that all scatter laser scars are located outside the macula. (UCSF Department of Ophthalmology) |
|
|
|
Figure 16. View of laser scars superior to the macula in the same eye. Spots are approximately one-half burn width apart. In the treated area, the retinal vessels are sclerotic (arrows). (UCSF Department of Ophthalmology) |
The Diabetic Retinopathy Study evaluated the effects of scatter photocoagulation in over 1700 patients with PDR or severe NPDR. Patients had one eye randomized to treatment and one eye to observation. Treatment was shown to reduce severe visual loss by 50%.(7) The ETDRS also found a positive risk-benefit ratio for early scatter treatment in patients with severe NPDR or early PDR.(6)
Surgery may be necessary for eyes in advanced PDR with either vitreous hemorrhage or retinal detachment. In the case of vitreous hemorrhage, many cases will clear spontaneously. For this reason, clinicians often wait 3 to 6 months or more before performing vitrectomy surgery. If surgery is indicated because of persistent nonclearing hemorrhage, retinal detachment involving the macula, or vitreous hemorrhage with neovascularization of the anterior chamber angle (a precursor of neovascular glaucoma), then vitrectomy is performed via a pars plana approach. The vitreous is removed, fibrovascular membranes are dissected away from the retina, retinal detachment is repaired, and scatter laser treatment is applied at the time of surgery via direct intraocular application.
The Diabetic Retinopathy Vitrectomy Study assessed the value of early vitrectomy in patients with severe PDR. The study found that early intervention increased the likelihood of obtaining 20/40 vision or better in eyes with recent severe vitreous hemorrhage or severe PDR. Compared with 15% of control eyes, 25% of treated eyes achieved this level of vision at 2 years.(6) In type 1 diabetes, the benefit of early surgery was even more pronounced, with 36% of treated eyes achieving 20/40 vision compared to 12% of control eyes. The importance of this study, performed between 1976 and 1983 when vitrectomy techniques were much less advanced than they are today, was that it showed conventional "watch and wait" management will not necessarily lead to the best visual outcomes in cases of severe PDR. In practice, clinicians evaluate the risks and benefits of each option before proceeding with scatter photocoagulation, vitrectomy, or observation in such cases.
TREATMENT: NOVEL THERAPIES FOR DIABETIC RETINOPATHY
Current therapies are limited in their ability to reverse vision loss in diabetic retinopathy. For example, although focal laser photocoagulation can help stabilize vision by reducing macular edema, it rarely improves vision. The development of new treatment modalities is being guided by an understanding of the mechanisms of the disease. Research on the inflammatory response in diabetic macular edema has led to studies evaluating the use of intravitreal corticosteroid injection, a promising treatment option in patients who do not benefit from laser.(13) Protein kinase C-ß inhibitors and antiangiogenic therapies targeted toward VEGF and other growth factors are the focus of study in animal models and early clinical trials.(14)
Retinopathy remains a challenging complication of diabetes that can adversely affect a patient's quality of life. Although ophthalmologists can often stabilize the condition or reduce vision loss, prevention and early detection remain the most effective ways to preserve good vision in patients with diabetes. Ensuring tight glucose and blood pressure control and referring patients for ophthalmologic examination are important ways in which internists and other clinicians can help to maximize their patients' vision and therefore their quality of life.
CLASSIFICATION OF DIABETES (Table 1)
Diabetes is a metabolic disorder characterized by resistance to the action of insulin, insufficient insulin secretion, or both (1). The major clinical manifestation of the diabetic state is hyperglycemia. However, insulin deficiency and/or insulin resistance also are associated with disturbances in lipid and protein metabolism. The vast majority of diabetic patients are classified into one of two broad categories: type 1 diabetes, which is caused by an absolute deficiency of insulin, and type 2 diabetes, which is characterized by the presence of insulin resistance with an inadequate compensatory increase in insulin secretion. In addition, women who develop diabetes during their pregnancy, are classified as having gestational diabetes. Finally, there are a variety of uncommon and diverse types of diabetes which are caused by infections, drugs, endocrinopathies, pancreatic destruction, and genetic defects. These unrelated forms of diabetes are classified separately.
|
Table 1: Etiologic Classification of Diabetes Mellitus |
|
I. Type 1 diabetes* (b-cell destruction, usually leading to absolute insulin deficiency)
|
|
II. Type 2 diabetes* (may range from predominantly insulin resistance with relative insulin deficiency to a predominantly insulin secretory defect with insulin resistance) |
|
III. Other specific types
|
|
IV. Gestational diabetes-melllitus (GDM) |
|
*Patients with any form of diabetes may require insulin treatment at some stage of their disease. Such use of insulin does not, of itself, classify the patient. Adapted from reference #29 with permission. |
Type 1 diabetes results from autoimmune destruction of the pancreatic b-cells (2,3). Markers of immune destruction of the b-cell are present at the time of diagnosis in 90% of individuals and include antibodies to the islet cell (ICAs), to glutamic acid decarboxylase (GAD), and to insulin (IAAs). While this form of diabetes usually occurs in children and adolescents, it can occur at any age. Younger individuals typically have a rapid rate of b -cell destruction and present with ketoacidosis, while adults often maintain sufficient insulin secretion to prevent ketoacidosis for many years (4). The more indolent adult-onset variety has been referred to as latent autoimmune diabetes in adults (LADA). Eventually, all type 1 diabetic patients will require insulin therapy to maintain normglycemia.
Type 2 diabetes is characterized by insulin resistance and, at least initially, a relative deficiency of insulin secretion (5,6). In absolute terms, the plasma insulin concentration (both fasting and meal-stimulated) usually is increased, although "relative" to the severity of insulin resistance, the plasma insulin concentration is insufficient to maintain normal glucose homeostasis (5,6). With time, however, there is progressive beta cell failure and absolute insulin deficiency ensues. In a minority of type 2 diabetic individuals, severe insulinopenia is present at the time of diagnosis and insulin sensitivity is normal or near normal (7). Most individuals with type 2 diabetes exhibit intra (abdominal (visceral) obesity, which is closely related to the presence of insulin resistance (4). In addition, hypertension, dyslipidemia (high triglyceride and low HDL-cholesterol levels; postprandial hyperlipemia), and elevated PAI-1 levels often are present in these individuals. This clustering of abnormalities is referred to as the "insulin resistance syndrome" or the "metabolic syndrome" (9,10). Because of these abnormalities, patients with type 2 diabetes are at increased risk of developing macrovascular complications (myocardial infarction and stroke). Type 2 diabetes has a strong genetic predisposition and is more common in minority ethnic groups, i.e. Mexican-Americans, Latinos, American Indians, Pacific Islanders, than in individuals of European ancestry. The genetic cause(s) of the common variety of type 2 diabetes is (are) not well defined and, at present, no specific genes have been identified in the pathogenesis of this common metabolic disorder (6,11).
GESTATIONAL DIABETES MELLITUS (GDM)
Gestational diabetes mellitus (GDM) is defined as glucose intolerance, which is first recognized during pregnancy. In most women who develop GDM, the disorder has its onset in the third trimester of pregnancy. At least 6 weeks after the pregnancy ends, the woman should receive an oral glucose tolerance test and be reclassified as having diabetes, normal glucose tolerance, impaired glucose tolerance, or impaired fasting glucose. Gestational diabetes complicates about 4% of all pregnancies (12). Clinical detection is important, since therapy will reduce perinatal morbidity and mortality. Risk assessment for GDM should occur at the first prenatal visit. Women at high risk (positive family history, history of GDM, marked obesity, high risk ethnic group) should be screened as soon as feasible. If the initial screening is negative, they should undergo retesting at 24-48 weeks. Women of average risk should have the initial screen performed at 24-48 weeks. A fasting plasma glucose concentration greater than 126 mg/dl (7.0 mmol/l) or a postprandial glucose greater than 200 mg/dl (11.1 mmol/l) establishes the diagnosis of diabetes and obviates the need for more formal glucose tolerance testing. Women who require more formal testing should receive a 100 gram oral glucose load with plasma glucose levels determined at baseline, 1 hour, 2 hours, and 3 hours (Table 2). The diagnosis of GDM is made if two or more of the plasma glucose values in Table 2 are met or exceeded.
|
Table 2-Diagnosis of GDM with a 100 g glucose load |
|
|
TIME |
PLASMA GLUCOSE |
|
Fasting |
≥95 mg/dl (5.3 mmol/L) |
|
1-h |
≥180 mg/dl (10.0mmol/L) |
|
2-h |
≥155 mg/dl (8.6 mmol/L) |
|
3-h |
≥140 mg/dl (7.8 mmol/L) |
|
Two or more values must be met or exceeded for a diagnosis of diabetes to be made. The test should be done in the morning after a 8 to 14 hour fast. |
|
Maturity Onset Diabetes of the Young (MODY) is characterized by impaired insulin secretion with minimal or no insulin resistance (13). Patients typically exhibit mild hyperglycemia at an early age. The disease is inherited in an autosomal dominant pattern and, at present, six different genetic abnormalities have been identified (14).
Genetic inability to convert proinsulin to insulin results in mild hyperglycemia and is inherited an autosomal dominant pattern (15). Similarly, the production of mutant insulin molecules has been identified in a few families and results in mild glucose intolerance (16).
Several genetic mutations have been described in the insulin receptor and are associated with insulin resistance (17). Type A insulin resistance refers to the clinical syndrome of acanthosis nigricans, virilization in women, polycystic ovaries, and hyperinsulinemia (18). Leprechaunism is a pediatric syndrome with specific facial features and severe insulin resistance that results from a defect in the insulin receptor (19). Lipoatrophic diabetes results from postreceptor defects in insulin signaling (20).
A variety of genetic syndromes have been described in which diabetes mellitus occurs with increased frequency. The etiology of the disturbance in glucose homeostasis in these diverse and seemingly unrelated syndromes remains undefined.
DISEASES OF THE EXOCRINE PANCREAS
Damage of the pancreas must be extensive for diabetes to occur (21). The most common causes are pancreatitis, trauma, and carcinoma. Cystic fibrosis and hemochromatosis also have been associated with impaired insulin secretion.
Since growth hormone, cortisol, glucagon, and epinephrine increase hepatic glucose production and induce insulin resistance in peripheral (muscle) tissues, excess production of these hormones can cause or exacerbate underlying diabetes (22-24). Although the primary mechanism of action of these counter regulatory hormones is the induction of insulin resistance in muscle and liver, overt type 2 diabetes mellitus does not develop in the absence of beta cell failure.
A variety of infections have been etiologically related to the development of diabetes mellitus. Of these, the most clearly established is congenital rubella (25). Approximately 20% of infants who are infected with the rubella virus at birth develop autoimmune type 2 diabetes later in life. These individuals have the typical type 1 susceptibility genotype, DR3/DR4.
A large number of commonly used drugs have been shown to induce insulin resistance and/or impair beta cell function and can lead to the development of diabetes mellitus in susceptible individuals. An extensive review of these drugs and their mechanism of action has been published (26).
The diagnosis of diabetes requires the identification of a glycemic cutpoint which discriminates normal individuals from those with diabetes. The present cutpoints reflect the level of glucose above which microvascular complications have been shown to increase. Cross-sectional studies from Egypt, Pima Indians, and a representative sample from the United States have shown a consistent increase in the risk of developing retinopathy when the fasting plasma glucose concentration exceeds 108-116 mg/dl (6.0-6.4 mmol/l), when the 2 hour postprandial level rises above 185 mg/dl (10.3 mmol/l), and when the hemoglobin A1c is greater then 5.9-6.0% (Figure 1) (27-29). Based upon these prospective epidemiologic studies relating glycemic control to the development of diabetic retinopathy, the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus of the ADA in 1997 revised the criteria for establishing the diagnosis of diabetes (29) (Table 3). To minimize the discrepancy between the fasting plasma glucose and 2-hour plasma glucose concentration measured during the OGTT, cut off values of ≥ 126 mg/dl and ≥ 200 mg/dl, respectively, were chosen. The WHO adopted this change in 1998 (30). Although the ADA recommended use of the fasting plasma glucose concentration as the principal tool for the diagnosis of diabetes mellitus in non-pregnant adults, the recently published results of the Diabetes Prevention Program (31) have given renewed emphasis to the OGTT, since diet/exercise, as well as drug therapy (metformin and troglitazone), were shown to slow/prevent the progression of IGT to overt diabetes mellitus. The diagnosis of IGT only can be made from the 2-hour plasma glucose concentration (≥ 140 to 199 mg/dl) during the OGTT. In addition the Expert Committee defined a new category of glycemia, impaired fasting glucose (IFG). IFG is defined by a plasma glucose ≥ 110 mg/dl (6.1 mmol/l) but less than 126 mg/dl (7.0 mmol/l). This category was created to correspond to the category of impaired glucose tolerance (IGT), which is defined as a 2 hour glucose value ≥140 mg/dl (7.8 mmol/l) but less than 200 mg/dl (11.0 mmol/l) during an OGTT.
|
|
|
Figure 1: Prevalence of retinopathy by deciles of the distribution FPG, 2-h PG, and HbA1c in Pima Indians (A) described in McCance et al (28), Egyptians (B) described in Engelgau et al (27), and in 40- to 74-year old participants in NHANES III (C) (K. Felgal, National Center for Health Statistics, as reported in reference 29). The x-axis labels indicate the lower limit of each decile group. Note that these deciles and the prevalence rates of retinopathy differ considerably among the studies, especially the Egyptian study, in which diabetic subjects were oversampled. Retinopathy was ascertained by different methods in each study; therefore, the absolute prevalence rates are not comparable between studies, but their relationships with FPG, 2-h, PG, and HbA1c are very similar within each population. |
|
Table 3. ADA criteria for the diagnosis of diabetes mellitus, impaired glucose tolerance (IGT), and impaired fasting glucose (IFG) |
|||
|
|
Diabetes |
IGT |
IFG |
|
FPG* |
≥126 |
<126 |
≥110 to 125 |
|
2-hour PG* |
≥200 |
≥140 and <200 |
<200 |
|
*FPG = fasting plasma glucose; 2-hour PG = 2-hour plasma glucose concentration during a standard oral glucose (75 gram) tolerance test |
|||
The fasting and postprandial glucose levels do not measure the same physiologic processes and, not surprisingly, they do not identify the same individuals as having diabetes. The fasting plasma glucose concentration is, in large part, determined by basal rate of hepatic glucose production (5,6) Thus, IFG primarily reflects hepatic resistance to the action of insulin. Under basal (postabsorptive) conditions, the majority of glucose is taken up by insulin-independent tissues (brain and liver) (5); although tissue (muscle) glucose clearance is reduced in the postabsorptive state, in absolute terms the muscle is responsible for only a small amount of glucose uptake in the basal state and is unlikely to explain the rise in fasting glucose concentration in individuals with IFG (5,6). Moreover, .basal insulin secretion is well preserved, even in individuals with overt type 2 diabetes mellitus (5), and, therefore, cannot explain the rise in fasting plasma glucose concentration in individuals with IFG. In contrast, the postprandial plasma glucose concentration primarily depends on insulin sensitivity in muscle and liver, as well as on insulin secretion by the pancreatic beta cells (5), and defects in both tissue (muscle) sensitivity to insulin and impaired insulin secretion are responsible for IGT. Although both IFG and IGT predict the future development of type 2 diabetes, IFG is a poor predictor of ASCVD, whereas IGT is a strong predictor of myocardial infarction and stroke (32). This discordance most likely reflects the association of IGT with the metabolic syndrome and insulin resistance in muscle (5,6,9). Although IFG and IGT are equally strong predictors of the development of future type 2 diabetes mellitus (32), the prevalence of IFG in the general population is significantly less than the prevalence of IGT (33). In order to have similar prevalences of the two disorders (IGF and IGT) of glucose homeostasis, the cutpoint for IFG would have to be reduced to approximately 103-104 mg/dl.
The use of the fasting plasma glucose concentration (≥ 126 mg/dl), as opposed to the 2-hour plasma glucose concentration during the OGTT (≥ 200 mg/dl), also significantly underestimates the prevalence of diabetes in the general population (32,33). In one study of U.S. adults between the ages of 40-74 years, the prevalence of undiagnosed diabetes was 6.4% using the OGTT and 4.4% based upon the fasting glucose measurement (34). These differences have been reported to be even greater in some ethnic groups (35-37).
The American Diabetes Association recommends use of hemoglobin A1c (HbA1c or A1c) determinations to monitor glycemic control in known diabetic patients. Because there is not a "gold standard" assay and because many countries do not have ready access to the test, an A1c determination is not recommended for the diagnosis of diabetes mellitus. However, because the A1c accurately reflects the mean blood glucose concentration over a 1-3 month period and correlates well with the development of diabetic complications, it may in the future become established as a test for the diagnosis of diabetes (38). The National Glycohemoglobin Standardization Program has established standard assays for A1c based on the results of the Diabetes Control and Complication Trials (39). In 1999, 78% of laboratories were using standardized assays with inter-laboratory coefficients of variation of less than 5% (40,41). An A1c level above ~6.5% correlates with the present diagnostic cutpoints for fasting and 2 hour postprandial plasma glucose levels.