Basic and Clinical Pharmacology, 13th Ed.

Pancreatic Hormones & Antidiabetic Drugs

Martha S. Nolte Kennedy, MD, & Umesh Masharani, MBBS, MRCP (UK)


A 56-year-old Hispanic woman presents to her medical practitioner with symptoms of fatigue, increased thirst, frequent urination, and exercise intolerance with shortness of breath of many months’ duration. She does not get regular medical care and is unaware of any medical problems. Her family history is significant for obesity, diabetes, high blood pressure, and coronary artery disease in both parents and several siblings. She is not taking any medications. Five of her six children had a birthweight of over 9 pounds. Physical examination reveals a BMI (body mass index) of 34, blood pressure of 150/90 mm Hg, and evidence of mild peripheral neuropathy. Laboratory tests reveal a random blood sugar of 261 mg/dL; this is confirmed with a fasting plasma glucose of 192 mg/dL. A fasting lipid panel reveals total cholesterol 264 mg/dL, triglycerides 255 mg/dL, high-density lipoproteins 43 mg/dL, and low-density lipoproteins 170 mg/dL. What type of diabetes does this woman have? What further evaluations should be obtained? How would you treat her diabetes?


The endocrine pancreas in the adult human consists of approximately 1 million islets of Langerhans interspersed throughout the pancreatic gland. Within the islets, at least five hormone-producing cells are present (Table 41–1). Their hormone products include insulin, the storage and anabolic hormone of the body; islet amyloid polypeptide (IAPP, or amylin), which modulates appetite, gastric emptying, and glucagon and insulin secretion; glucagon, the hyperglycemic factor that mobilizes glycogen stores; somatostatin, a universal inhibitor of secretory cells; pancreatic peptide, a small protein that facilitates digestive processes by a mechanism not yet clarified; and ghrelin, a peptide known to increase pituitary growth hormone release.

TABLE 41–1 Pancreatic islet cells and their secretory products.


Diabetes mellitus is defined as an elevated blood glucose associated with absent or inadequate pancreatic insulin secretion, with or without concurrent impairment of insulin action. The disease states underlying the diagnosis of diabetes mellitus are now classified into four categories: type 1, type 2, other, and gestational diabetes mellitus.

Type 1 Diabetes Mellitus

The hallmark of type 1 diabetes is selective beta cell (B cell) destruction and severe or absolute insulin deficiency. Type 1 diabetes is further subdivided into immune-mediated (type 1a) and idiopathic causes (type 1b). The immune form is the most common form of type 1 diabetes. Although most patients are younger than 30 years of age at the time of diagnosis, the onset can occur at any age. Type 1 diabetes is found in all ethnic groups, but the highest incidence is in people from northern Europe and from Sardinia. Susceptibility appears to involve a multifactorial genetic linkage, but only 10–15% of patients have a positive family history. Most patients with type 1 diabetes have one or more circulating antibodies to glutamic acid decarboxylase 65 (GAD 65), insulin autoantibody, tyrosine phosphatase IA2 (ICA 512), and zinc transporter 8 (ZnT8) at the time of diagnosis. These antibodies facilitate the diagnosis of type 1a diabetes and can also be used to screen family members at risk for developing the disease.

For persons with type 1 diabetes, insulin replacement therapy is necessary to sustain life. Pharmacologic insulin is administered by injection into the subcutaneous tissue using a manual injection device or an insulin pump that continuously infuses insulin under the skin. Interruption of the insulin replacement therapy can be life-threatening and can result in diabetic ketoacidosis or death. Diabetic ketoacidosis is caused by insufficient or absent insulin and results from excess release of fatty acids and subsequent formation of toxic levels of ketoacids.

Some patients with type 1 diabetes have a more indolent autoimmune process and initially retain enough beta cell function to avoid ketosis. They can be treated at first with oral hypoglycemic agents but then need insulin as their beta cell function declines. Antibody studies in northern Europeans indicate that up to 10–15% of “type 2” patients may actually have this milder form of type 1 diabetes (latent autoimmune diabetes of adulthood; LADA).

Type 2 Diabetes Mellitus

Type 2 diabetes is characterized by tissue resistance to the action of insulin combined with a relative deficiency in insulin secretion. A given individual may have more resistance or more beta-cell deficiency, and the abnormalities may be mild or severe. Although insulin is produced by the beta cells in these patients, it is inadequate to overcome the resistance, and the blood glucose rises. The impaired insulin action also affects fat metabolism, resulting in increased free fatty acid flux and triglyceride levels and reciprocally low levels of high-density lipoprotein (HDL).

Individuals with type 2 diabetes may not require insulin to survive, but 30% or more will benefit from insulin therapy to control blood glucose. Although persons with type 2 diabetes ordinarily do not develop ketosis, ketoacidosis may occur as the result of stress such as infection or the use of medication that enhances resistance, eg, corticosteroids. Dehydration in individuals with untreated or poorly controlled type 2 diabetes can lead to a life-threatening condition called nonketotic hyperosmolar coma. In this condition, the blood glucose may rise to 6–20 times the normal range and an altered mental state develops or the person loses consciousness. Urgent medical care and rehydration are required.

Other Specific Types of Diabetes Mellitus

The “other” designation refers to multiple other specific causes of an elevated blood glucose: pancreatectomy, pancreatitis, nonpan-creatic diseases, drug therapy, etc. For a detailed list the reader is referred to the reference Expert Committee, 2003.

Gestational Diabetes Mellitus

Gestational diabetes (GDM) is defined as any abnormality in glucose levels noted for the first time during pregnancy. Gestational diabetes is diagnosed in approximately 7% of all pregnancies in the USA. During pregnancy, the placenta and placental hormones create an insulin resistance that is most pronounced in the last trimester. Risk assessment for diabetes is suggested starting at the first prenatal visit. High-risk women should be screened immediately. Screening may be deferred in lower-risk women until the 24th to 28th week of gestation.



Insulin is a small protein with a molecular weight in humans of 5808. It contains 51 amino acids arranged in two chains (A and B) linked by disulfide bridges; there are species differences in the amino acids of both chains. Proinsulin, a long single-chain protein molecule, is processed within the Golgi apparatus of beta cells and packaged into granules, where it is hydrolyzed into insulin and a residual connecting segment called C-peptide by removal of four amino acids (Figure 41–1).


FIGURE 41–1 Structure of human proinsulin (C-peptide plus A and B chains) and insulin. Insulin is shown as the shaded (orange color) peptide chains, A and B. Differences in the A and B chains and amino acid modifications for the rapid-acting insulin analogs (aspart, lispro, and glulisine) and long-acting insulin analogs (glargine and detemir) are discussed in the text. (Adapted, with permission, from Gardner DG, Shoback D [editors]: Greenspan’s Basic & Clinical Endocrinology, 9th ed. McGraw-Hill, 2011. Copyright © The McGraw-Hill Companies, Inc.)

Insulin and C-peptide are secreted in equimolar amounts in response to all insulin secretagogues; a small quantity of unprocessed or partially hydrolyzed proinsulin is released as well. Although proinsulin may have some mild hypoglycemic action, C-peptide has no known physiologic function. Granules within the beta cells store the insulin in the form of crystals consisting of two atoms of zinc and six molecules of insulin. The entire human pancreas contains up to 8 mg of insulin, representing approximately 200 biologic units. Originally, the unit was defined on the basis of the hypoglycemic activity of insulin in rabbits. With improved purification techniques, the unit is presently defined on the basis of weight, and present insulin standards used for assay purposes contain 28 units per milligram.

Insulin Secretion

Insulin is released from pancreatic beta cells at a low basal rate and at a much higher stimulated rate in response to a variety of stimuli, especially glucose. Other stimulants such as other sugars (eg, mannose), amino acids (especially gluconeogenic amino acids, eg, leucine, arginine), hormones such as glucagon-like polypeptide-1 (GLP-1), glucose-dependent insulinotropic polypeptide (GIP), glucagon, cholecystokinin, high concentrations of fatty acids, and β-adrenergic sympathetic activity are recognized. Stimulatory drugs are sulfonylureas, meglitinide and nateglinide, isoproterenol, and acetylcholine. Inhibitory signals are hormones including insulin itself, somatostatin, and leptin; α-adrenergic sympathetic activity; chronically elevated glucose; and low concentrations of fatty acids. Inhibitory drugs include diazoxide, phenytoin, vinblastine, and colchicine.

One mechanism of stimulated insulin release is diagrammed in Figure 41–2. As shown in the figure, hyperglycemia results in increased intracellular ATP levels, which close the ATP-dependent potassium channels. Decreased outward potassium efflux results in depolarization of the beta cell and opening of voltage-gated calcium channels. The resulting increased intracellular calcium triggers secretion of the hormone. The insulin secretagogue drug group (sulfonylureas, meglitinides, and D-phenylalanine) exploits parts of this mechanism.


FIGURE 41–2 One model of control of insulin release from the pancreatic beta cell by glucose and by sulfonylurea drugs. In the resting cell with normal (low) ATP levels, potassium diffuses down its concentration gradient through ATP-gated potassium channels, maintaining the intracellular potential at a fully polarized, negative level. Insulin release is minimal. If glucose concentration rises, ATP production increases, potassium channels close, and depolarization of the cell results. As in muscle and nerve, voltage-gated calcium channels open in response to depolarization, allowing more calcium to enter the cell. Increased intracellular calcium results in increased insulin secretion. Insulin secretagogues close the ATP-dependent potassium channel, thereby depolarizing the membrane and causing increased insulin release by the same mechanism.

Insulin Degradation

The liver and kidney are the two main organs that remove insulin from the circulation. The liver normally clears the blood of approximately 60% of the insulin released from the pancreas by virtue of its location as the terminal site of portal vein blood flow, with the kidney removing 35–40% of the endogenous hormone. However, in insulin-treated diabetics receiving subcutaneous insulin injections, this ratio is reversed, with as much as 60% of exogenous insulin being cleared by the kidney and the liver removing no more than 30–40%. The half-life of circulating insulin is 3–5 minutes.

Circulating Insulin

Basal serum insulin values of 5–15 μU/mL (30–90 pmol/L) are found in normal humans, with a peak rise to 60–90 μU/mL (360–540 pmol/L) during meals.

The Insulin Receptor

After insulin has entered the circulation, it diffuses into tissues, where it is bound by specialized receptors that are found on the membranes of most tissues. The biologic responses promoted by these insulin-receptor complexes have been identified in the primary target tissues regulating energy metabolism, ie, liver, muscle, and adipose tissue. The receptors bind insulin with high specificity and affinity in the picomolar range. The full insulin receptor consists of two covalently linked heterodimers, each containing an α subunit, which is entirely extracellular and constitutes the recognition site, and a β subunit that spans the membrane (Figure 41–3). The β subunit contains a tyrosine kinase. The binding of an insulin molecule to the α subunits at the outside surface of the cell activates the receptor and through a conformational change brings the catalytic loops of the opposing cytoplasmic β subunits into closer proximity. This facilitates mutual phosphorylation of tyrosine residues on the β subunits and tyrosine kinase activity directed at cytoplasmic proteins.


FIGURE 41–3 Schematic diagram of the insulin receptor heterodimer in the activated state. IRS, insulin receptor substrate; MAP, mitogen-activated protein; P, phosphate; Tyr, tyrosine.

The first proteins to be phosphorylated by the activated receptor tyrosine kinases are the docking proteins, insulin receptor substrates (IRS). After tyrosine phosphorylation at several critical sites, the IRS molecules bind to and activate other kinases subserving energy metabolism—most significantly phosphatidylinositol-3-kinase—which produce further phosphorylations. Alternatively, they may stimulate a mitogenic pathway and bind to an adaptor protein such as growth factor receptor-binding protein 2, which translates the insulin signal to a guanine nucleotide-releasing factor that ultimately activates the GTP binding protein, Ras, and the mitogen-activated protein kinase (MAPK) system. The particular IRS-phosphorylated tyrosine kinases have binding specificity with downstream molecules based on their surrounding 4–5 amino acid sequences or motifs that recognize specific Src homology 2 (SH2) domains on the other protein. This network of phosphorylations within the cell represents insulin’s second message and results in multiple effects, including translocation of glucose transporters (especially GLUT 4, Table 41–2) to the cell membrane with a resultant increase in glucose uptake; increased glycogen synthase activity and increased glycogen formation; multiple effects on protein synthesis, lipolysis, and lipogenesis; and activation of transcription factors that enhance DNA synthesis and cell growth and division.

TABLE 41–2 Glucose transporters.


Various hormonal agents (eg, glucocorticoids) lower the affinity of insulin receptors for insulin; growth hormone in excess increases this affinity slightly. Aberrant serine and threonine phosphorylation of the insulin receptor β subunits or IRS molecules may result in insulin resistance and functional receptor down-regulation.

Effects of Insulin on Its Targets

Insulin promotes the storage of fat as well as glucose (both sources of energy) within specialized target cells (Figure 41–4) and influences cell growth and the metabolic functions of a wide variety of tissues (Table 41–3).


FIGURE 41–4 Insulin promotes synthesis (from circulating nutrients) and storage of glycogen, triglycerides, and protein in its major target tissues: liver, fat, and muscle. The release of insulin from the pancreas is stimulated by increased blood glucose, incretins, vagal nerve stimulation, and other factors (see text).

TABLE 41–3 Endocrine effects of insulin.


Characteristics of Available Insulin Preparations

Commercial insulin preparations differ in a number of ways, such as differences in the recombinant DNA production techniques, amino acid sequence, concentration, solubility, and the time of onset and duration of their biologic action.

A. Principal Types and Duration of Action of Insulin Preparations

Four principal types of injected insulins are available: (1) rapid-acting, with very fast onset and short duration; (2) short-acting, with rapid onset of action; (3) intermediate-acting; and (4) long-acting, with slow onset of action (Figure 41–5Table 41–4). Injected rapid-acting and short-acting insulins are dispensed as clear solutions at neutral pH and contain small amounts of zinc to improve their stability and shelf life. Injected intermediate-acting NPH insulins have been modified to provide prolonged action and are dispensed as a turbid suspension at neutral pH with protamine in phosphate buffer (neutral protamine Hagedorn [NPH] insulin). Insulin glargine and insulin detemir are clear, soluble long-acting insulins.


FIGURE 41–5 Extent and duration of action of various types of insulin as indicated by the glucose infusion rates (mg/kg/min) required to maintain a constant glucose concentration. The durations of action shown are typical of an average dose of 0.2–0.3 U/kg. The durations of regular and NPH insulin increase considerably when dosage is increased.

TABLE 41–4 Some insulin preparations available in the USA.1


The goal of subcutaneous insulin therapy is to replicate normal physiologic insulin secretion and replace the background or basal (overnight, fasting, and between-meal) as well as bolus or prandial (mealtime) insulin. An exact reproduction of the normal glycemic profile is not technically possible because of the limitations inherent in subcutaneous administration of insulin. Current regimens generally use insulin analogs because of their more predictable action. Intensive therapy (“tight control”) attempts to restore near-normal glucose patterns throughout the day while minimizing the risk of hypoglycemia.

Intensive regimens involving multiple daily injections (MDI) use long-acting insulin analogs to provide basal or background coverage, and rapid-acting insulin analogs to meet the mealtime requirements. The latter insulins are given as supplemental doses to correct transient hyperglycemia. The most sophisticated insulin regimen delivers rapid-acting insulin analogs through a continuous subcutaneous insulin infusion device. Conventional therapy consists of split-dose injections of mixtures of rapid- or short-acting and intermediate-acting insulins.

1. Rapid-acting insulinThree injected rapid-acting insulin analogs—insulin lispro, insulin aspart, and insulin glulisine—are commercially available. The rapid-acting insulins permit more physiologic prandial insulin replacement because their rapid onset and early peak action more closely mimic normal endogenous prandial insulin secretion than regular insulin, and they have the additional benefit of allowing insulin to be taken immediately before the meal without sacrificing glucose control. Their duration of action is rarely more than 4–5 hours, which decreases the risk of late postmeal hypoglycemia. The injected rapid-acting insulins have the lowest variability of absorption (approximately 5%) of all available commercial insulins (compared with 25% for regular insulin and 25% to over 50% for long-acting analog formulations and intermediate insulin, respectively). They are the preferred insulins for use in continuous subcutaneous insulin infusion devices.

Insulin lispro, the first monomeric insulin analog to be marketed, is produced by recombinant technology wherein two amino acids near the carboxyl terminal of the B chain have been reversed in position: Proline at position B28 has been moved to B29, and lysine at position B29 has been moved to B28 (Figure 41–1). Reversing these two amino acids does not interfere in any way with insulin lispro’s binding to the insulin receptor, its circulating half-life, or its immunogenicity, which are similar to those of human regular insulin. However, the advantage of this analog is its very low propensity—in contrast to human insulin—to self-associate in antiparallel fashion and form dimers. To enhance the shelf life of insulin in vials, insulin lispro is stabilized into hexamers by a cresol preservative. When injected subcutaneously, the drug quickly dissociates into monomers and is rapidly absorbed with onset of action within 5–15 minutes and peak activity as early as 1 hour. The time to peak action is relatively constant, regardless of the dose.

Insulin aspart is created by the substitution of the B28 proline with a negatively charged aspartic acid (Figure 41–1). This modification reduces the normal ProB28 and GlyB23 monomer-monomer interaction, thereby inhibiting insulin self-aggregation. Its absorption and activity profile are similar to those of insulin lispro, and it is more reproducible than regular insulin, but it has binding properties, activity, and mitogenicity characteristics similar to those of regular insulin in addition to equivalent immunogenicity.

Insulin glulisine is formulated by substituting a lysine for asparagine at B3 and glutamic acid for lysine at B29. Its absorption, action, and immunologic characteristics are similar to those of other injected rapid-acting insulins. After high-dose insulin glulisine interaction with the insulin receptor, there may be downstream differences in IRS-2 pathway activation relative to native insulin. The clinical significance of such differences is unclear.

2. Short-acting insulinRegular insulin is a short-acting soluble crystalline zinc insulin that is now made by recombinant DNA techniques to produce a molecule identical to human insulin. Its effect appears within 30 minutes, peaks between 2 and 3 hours after subcutaneous injection, and generally lasts 5–8 hours. In high concentrations, eg, in the vial, regular insulin molecules self-aggregate in antiparallel fashion to form dimers that stabilize around zinc ions to create insulin hexamers. The hexameric nature of regular insulin causes a delayed onset and prolongs the time to peak action. After subcutaneous injection, the insulin hexamers are too large and bulky to be transported across the vascular endothelium into the bloodstream. As the insulin depot is diluted by interstitial fluid and the concentration begins to fall, the hexamers break down into dimers and finally monomers. This results in three rates of absorption of the injected insulin, with the final monomeric phase having the fastest uptake out of the injection site.

The clinical consequence is that when regular insulin is administered at mealtime, the blood glucose rises faster than the insulin with resultant early postprandial hyperglycemia and an increased risk of late postprandial hypoglycemia. Therefore, regular insulin should be injected 30–45 or more minutes before the meal to minimize the mismatching. As with all older insulin formulations, the duration of action as well as the time of onset and the intensity of peak action increase with the size of the dose. Clinically, this is a critical issue because the pharmacokinetics and pharmacodynamics of small doses of regular and NPH insulins differ greatly from those of large doses. The delayed absorption, dose-dependent duration of action, and variability of absorption (˜ 25%) of regular human insulin frequently results in a mismatching of insulin availability with need, and its use is declining.

However, short-acting, regular soluble insulin is the only type that should be administered intravenously because the dilution causes the hexameric insulin to immediately dissociate into monomers. It is particularly useful for intravenous therapy in the management of diabetic ketoacidosis and when the insulin requirement is changing rapidly, such as after surgery or during acute infections.

3. Intermediate-acting and long-acting insulins

a. NPH (neutral protamine Hagedorn, or isophane) insulin—NPH insulin is an intermediate-acting insulin whose absorption and onset of action are delayed by combining appropriate amounts of insulin and protamine so that neither is present in an uncomplexed form (“isophane”). After subcutaneous injection, proteolytic tissue enzymes degrade the protamine to permit absorption of insulin. NPH insulin has an onset of approximately 2–5 hours and duration of 4–12 hours (Figure 41–5); it is usually mixed with regular, lispro, aspart, or glulisine insulin and given two to four times daily for insulin replacement. The dose regulates the action profile; specifically, small doses have lower, earlier peaks and a short duration of action with the converse true for large doses. The action of NPH is highly unpredictable, and its variability of absorption is over 50%. The clinical use of NPH is waning because of its adverse pharmacokinetics combined with the availability of long-acting insulin analogs that have a more predictable and physiologic action.

b. Insulin glargine—Insulin glargine is a soluble, “peakless” (ie, having a broad plasma concentration plateau), long-acting insulin analog. This product was designed to provide reproducible, convenient, background insulin replacement. The attachment of two arginine molecules to the B-chain carboxyl terminal and substitution of a glycine for asparagine at the A21 position created an analog that is soluble in an acidic solution but precipitates in the more neutral body pH after subcutaneous injection. Individual insulin molecules slowly dissolve away from the crystalline depot and provide a low, continuous level of circulating insulin. Insulin glargine has a slow onset of action (1–1.5 hours) and achieves a maximum effect after 4–6 hours. This maximum activity is maintained for 11–24 hours or longer. Glargine is usually given once daily, although some very insulin-sensitive or insulin-resistant individuals benefit from split (twice a day) dosing. To maintain solubility, the formulation is unusually acidic (pH 4.0), and insulin glargine should not be mixed with other insulins. Separate syringes must be used to minimize the risk of contamination and subsequent loss of efficacy. The absorption pattern of insulin glargine appears to be independent of the anatomic site of injection, and this drug is associated with less immunogenicity than human insulin in animal studies. Glargine’s interaction with the insulin receptor is similar to that of native insulin and shows no increase in mitogenic activity in vitro. It has sixfold to sevenfold greater binding than native insulin to the insulin-like growth factor-1 (IGF-1) receptor, but the clinical significance of this is unclear.

c. Insulin detemir—This insulin is the most recently developed long-acting insulin analog. The terminal threonine is dropped from the B30 position and myristic acid (a C-14 fatty acid chain) is attached to the terminal B29 lysine. These modifications prolong the availability of the injected analog by increasing both self-aggregation in subcutaneous tissue and reversible albumin binding. Insulin detemir has the most reproducible effect of the intermediate- and long-acting insulins, and its use is associated with less hypoglycemia than NPH insulin. Insulin detemir has a dose-dependent onset of action of 1–2 hours and duration of action of more than 12 hours. It is given twice daily to obtain a smooth background insulin level.

4. Mixtures of insulinsBecause intermediate-acting NPH insulins require several hours to reach adequate therapeutic levels, their use in diabetic patients usually requires supplements of rapid-or short-acting insulin before meals. For convenience, these are often mixed together in the same syringe before injection. Insulin lispro, aspart, and glulisine can be acutely mixed (ie, just before injection) with NPH insulin without affecting their rapid absorption. However, premixed preparations have thus far been unstable. To remedy this, intermediate insulins composed of isophane complexes of protamine with insulin lispro and insulin aspart have been developed. These intermediate insulins have been designated as “NPL” (neutral protamine lispro) and “NPA” (neutral protamine aspart) and have the same duration of action as NPH insulin. They have the advantage of permitting formulation as premixed combinations of NPL and insulin lispro, and as NPA and insulin aspart, and they have been shown to be safe and effective in clinical trials. The FDA has approved 75%/25% NPL/insulin lispro and 70%/30% NPA/insulin aspart premixed formulations. Additional ratios are available abroad. Insulin glargine and detemir must be given as separate injections. They are not miscible acutely or in a premixed preparation with any other insulin formulation.

Premixed formulations of 70%/30% NPH/regular continue to be available. These preparations have all the limitations of regular insulin, namely, highly dose-dependent pharmacokinetic and pharmacodynamic profiles, and variability in absorption.

B. Insulin Production

Mass production of human insulin and insulin analogs by recombinant DNA techniques is carried out by inserting the human or a modified human proinsulin gene into Escherichia coli or yeast and treating the extracted proinsulin to form the insulin or insulin analog molecules.

C. Concentration

All insulins in the USA and Canada are available in a concentration of 100 U/mL (U100). A limited supply of U500 regular human insulin is available for use in rare cases of severe insulin resistance in which larger doses of insulin are required.

Insulin Delivery Systems

A. Standard Delivery

The standard mode of insulin therapy is subcutaneous injection using conventional disposable needles and syringes.

B. Portable Pen Injectors

To facilitate multiple subcutaneous injections of insulin, particularly during intensive insulin therapy, portable pen-sized injectors have been developed. These contain cartridges of insulin and replaceable needles.

Disposable insulin pens are also available for selected formulations. These are regular insulin, insulin lispro, insulin aspart, insulin glulisine, insulin glargine, insulin detemir, and several mixtures of NPH with regular, lispro, or aspart insulin (Table 41–4). They have been well accepted by patients because they eliminate the need to carry syringes and bottles of insulin to the workplace and while traveling.

C. Continuous Subcutaneous Insulin Infusion Devices (CSII, Insulin Pumps)

Continuous subcutaneous insulin infusion devices are external open-loop pumps for insulin delivery. The devices have a user-programmable pump that delivers individualized basal and bolus insulin replacement doses based on blood glucose self-monitoring results.

Normally, the 24-hour background basal rates are preprogrammed and relatively constant from day to day, although temporarily altered rates can be superimposed to adjust for a short-term change in requirement. For example, the basal delivery rate might need to be decreased for several hours because of the increased insulin sensitivity associated with strenuous activity.

Boluses are used to correct high blood glucose levels and to cover mealtime insulin requirements based on the carbohydrate content of the food and concurrent activity. Bolus amounts are either dynamically programmed or use preprogrammed algorithms. When the boluses are dynamically programmed, the user calculates the dose based on the amount of carbohydrate consumed and the current blood glucose level. Alternatively, the meal or snack dose algorithm (grams of carbohydrate covered by a unit of insulin) and insulin sensitivity or blood glucose correction factor (fall in blood glucose level in response to a unit of insulin) can be preprogrammed into the pump. If the user enters the carbohydrate content of the food and current blood glucose value, the insulin pump will calculate the most appropriate dose of insulin. Advanced insulin pumps also have an “insulin on board” feature that adjusts a high blood glucose correction dose to correct for residual activity of previous bolus doses.

The traditional pump—which contains an insulin reservoir, the program chip, the keypad, and the display screen—is about the size of a pager. It is usually placed on a belt or in a pocket, and the insulin is infused through thin plastic tubing that is connected to the subcutaneously inserted infusion set. The abdomen is the favored site for the infusion set, although flanks and thighs are also used. The insulin reservoir, tubing, and infusion set need to be changed using sterile techniques every 2 or 3 days. Currently, only one pump does not require tubing. In this model, the pump is attached directly to the infusion set. Programming is done through a hand-held unit that communicates wirelessly with the pump. CSII delivery is regarded as the most physiologic method of insulin replacement.

Use of these continuous infusion devices is encouraged for people who are unable to obtain target control while on multiple injection regimens and in circumstances in which excellent glycemic control is desired, such as during pregnancy. Optimal use of these devices requires responsible involvement and commitment by the patient. Insulin aspart, lispro, and glulisine all are specifically approved for pump use and are preferred pump insulins because their favorable pharmacokinetic attributes allow glycemic control without increasing the risk of hypoglycemia.

D. Inhaled Insulin

A dry powder formulation of recombinant regular insulin (technosphere insulin, Afrezza) is now approved for use in adults with diabetes. After inhalation from the small, single use device, peak levels are reached in 12 -15 minutes and decline to baseline in 3 hours, significantly faster in onset and shorter in duration than subcutaneous insulin. In trials, inhaled insulin combined with injected basal insulin was as effective in lowering glucose as injected rapid-acting insulin combined with basal insulin. The most common adverse effect of inhaled insulin was cough, affecting 27% of trial patients and pulmonary function should be monitored. The drug is contraindicated in smokers and patients with chronic obstructive pulmonary disease.

Treatment with Insulin

The current classification of diabetes mellitus identifies a group of patients who have virtually no insulin secretion and whose survival depends on administration of exogenous insulin. This insulin-dependent group (type 1) represents 10–15% of the diabetic population in the USA. Most type 2 diabetics do not require exogenous insulin for survival, but many need exogenous supplementation of their endogenous secretion to achieve optimum health.

Benefit of Glycemic Control in Diabetes Mellitus

The consensus of the American Diabetes Association is that intensive glycemic control and targeting normal or near-normal glucose control associated with comprehensive self-management training should become standard therapy in diabetic patients (see Box: Benefits of Tight Glycemic Control in Diabetes). Exceptions include patients with advanced renal disease and the elderly, because the risks of hypoglycemia may outweigh the benefit of normal or near-normal glycemic control in these groups. In children under 7 years, the extreme susceptibility of the developing brain to incur damage from hypoglycemia contraindicates attempts at intensive glycemic control.

Insulin Regimens

A. Intensive Insulin Therapy

Intensive insulin regimens are prescribed for almost everyone with type 1 diabetes—diabetes associated with a severe deficiency or absence of endogenous insulin production—as well as many with type 2 diabetes.

Generally, the total daily insulin requirement in units is equal to the weight in pounds divided by four, or 0.55 times the person’s weight in kilograms. Approximately half the total daily insulin dosage covers the background or basal insulin requirements, and the remainder covers meal and snack requirement and high blood sugar corrections. This is an approximate calculation and has to be individualized. Examples of reduced insulin requirement include newly diagnosed persons and those with ongoing endogenous insulin production, longstanding diabetes with insulin sensitivity, significant renal insufficiency, or other endocrine deficiencies. Increased insulin requirements typically occur with obesity, during adolescence, during the latter trimesters of pregnancy, and in individuals with type 2 diabetes.

In intensive insulin regimens, the meal or snack and high blood sugar correction boluses are prescribed by formulas. The patient uses the formulas to calculate the rapid-acting insulin bolus dose by considering how much carbohydrate is in the meal or snack, the current plasma glucose, and the target glucose. The formula for the meal or snack bolus is expressed as an insulin-to-carbohydrate ratio, which refers to how many grams of carbohydrate will be disposed of by 1 unit of rapid-acting insulin. The high blood sugar correction formula is expressed as the predicted fall in plasma glucose (in mg/dL) after 1 unit of rapid-acting insulin. Diurnal variations in insulin sensitivity can be accommodated by prescribing different basal rates and bolus insulin doses throughout the day. Continuous subcutaneous insulin infusion devices provide the most sophisticated and physiologic insulin replacement.

B. Conventional Insulin Therapy

Conventional insulin therapy is usually prescribed only for certain people with type 2 diabetes who are felt not to benefit from intensive glucose control. The insulin regimen ranges from one injection per day to many injections per day, using intermediate- or long-acting insulin alone or with short- or rapid-acting insulin or premixed insulins. Referred to as sliding-scale regimens, conventional insulin regimens customarily fix the dose of the intermediate- or long-acting insulin, but vary the short- or rapid-acting insulin based on the plasma glucose level before the injection. This insulin replacement regimen assumes similar daily timing and carbohydrate content of meals

Insulin Treatment of Special Circumstances

A. Diabetic Ketoacidosis

Diabetic ketoacidosis (DKA) is a life-threatening medical emergency caused by inadequate or absent insulin replacement, which occurs in people with type 1 diabetes and infrequently in those with type 2 diabetes. It typically occurs in newly diagnosed type 1 patients or in those who have experienced interrupted insulin replacement, and rarely in people with type 2 diabetes who have concurrent unusually stressful conditions such as sepsis or pancreatitis or are on high-dose steroid therapy. Signs and symptoms include nausea, vomiting, abdominal pain, deep slow (Kussmaul) breathing, change in mental status, elevated blood and urinary ketones and glucose, an arterial blood pH lower than 7.3, and low bicarbonate (< 15 mmol/L).

The fundamental treatment for DKA includes aggressive intravenous hydration and insulin therapy and maintenance of potassium and other electrolyte levels. Fluid and insulin therapy is based on the patient’s individual needs and requires frequent reevaluation and modification. Close attention has to be given to hydration and renal status, the sodium and potassium levels, and the rate of correction of plasma glucose and plasma osmolality. Fluid therapy generally begins with normal saline. Regular human insulin should be used for intravenous therapy with a usual starting dosage of about 0.1 U/kg/h.

B. Hyperosmolar Hyperglycemic Syndrome

Hyperosmolar hyperglycemic syndrome (HHS) is diagnosed in persons with type 2 diabetes and is characterized by profound hyperglycemia and dehydration. It is associated with inadequate oral hydration, especially in elderly patients, with other illnesses, the use of medication that elevates the blood sugar or causes dehydration, such as phenytoin, steroids, diuretics, and β blockers, and with peritoneal dialysis and hemodialysis. The diagnostic hallmarks are declining mental status and even seizures, a plasma glucose of over 600 mg/dL, and a calculated serum osmolality higher than 320 mmol/L. Persons with HHS are not acidotic unless DKA is also present.

Benefits of Tight Glycemic Control in Diabetes

A long-term randomized prospective study involving 1441 type 1 patients in 29 medical centers reported in 1993 that “near normalization” of blood glucose resulted in a delay in onset and a major slowing of progression of microvascular and neuropathic complications of diabetes during follow-up periods of up to 10 years (Diabetes Control and Complications Trial [DCCT] Research Group, 1993). In the intensively treated group, mean glycated hemoglobin (HbA1c) of 7.2% (normal < 6%) and mean blood glucose of 155 mg/dL were achieved, whereas in the conventionally treated group, HbA1c averaged 8.9% with mean blood glucose of 225 mg/dL. Over the study period, which averaged 7 years, a reduction of approximately 60% in risk of diabetic retinopathy, nephropathy, and neuropathy was noted in the tight control group compared with the standard control group.

The DCCT study, in addition, introduced the concept of glycemic memory, which comprises the long-term benefits of any significant period of glycemic control. During a 6-year follow-up period, both the intensively and the conventionally treated groups had similar levels of glycemic control, and both had progression of carotid intimal-medial thickness. However, the intensively treated cohort had significantly less progression of intimal thickness.

The United Kingdom Prospective Diabetes Study (UKPDS) was a very large randomized prospective study carried out to study the effects of intensive glycemic control with several types of therapies and the effects of blood pressure control in type 2 diabetic patients. A total of 3867 newly diagnosed type 2 diabetic patients were studied over 10 years. A significant fraction of these were overweight and hypertensive. Patients were given dietary treatment alone or intensive therapy with insulin, chlorpropamide, glyburide, or glipizide. Metformin was an option for patients with inadequate response to other therapies. Tight control of blood pressure was added as a variable, with an angiotensin-converting enzyme inhibitor, a β blocker, or in some cases, a calcium channel blocker available for this purpose.

Tight control of diabetes, with reduction of HbA1c from 9.1% to 7%, was shown to reduce the risk of microvascular complications overall compared with that achieved with conventional therapy (mostly diet alone, which decreased HbA1c to 7.9%). Cardiovascular complications were not noted for any particular therapy; metformin treatment alone reduced the risk of macrovascular disease (myocardial infarction, stroke). Epidemiologic analysis of the study suggested that every 1% decrease in the HbA1c achieved an estimated risk reduction of 37% for microvascular complications, 21% for any diabetes-related end point and death related to diabetes, and 14% for myocardial infarction.

Tight control of hypertension also had a surprisingly significant effect on microvascular disease (as well as more conventional hypertension-related sequelae) in these diabetic patients. Epidemiologic analysis of the results suggested that every 10 mm Hg decrease in the systolic pressure achieved an estimated risk reduction of 13% for diabetic microvascular complications, and 12% for any diabetes-related complication, 15% for death related to diabetes, and 11% for myocardial infarction.

Post-study monitoring showed that 5 years after the closure of the UKPDS, the benefits of intensive management on diabetic end points was maintained and the risk reduction for a myocardial infarction became significant. The benefits of metformin therapy were maintained.

These studies show that tight glycemic control benefits both type 1 and type 2 patients.

The treatment of HHS centers around aggressive rehydration and restoration of glucose and electrolyte homeostasis; the rate of correction of these variables must be monitored closely. Low-dose insulin therapy may be required.

Complications of Insulin Therapy

A. Hypoglycemia

1. Mechanisms and diagnosisHypoglycemic reactions are the most common complication of insulin therapy. They usually result from inadequate carbohydrate consumption, unusual physical exertion, or too large a dose of insulin.

Rapid development of hypoglycemia in persons with intact hypoglycemic awareness causes signs of autonomic hyperactivity—both sympathetic (tachycardia, palpitations, sweating, tremulousness) and parasympathetic (nausea, hunger)—and may progress to convulsions and coma if untreated.

In persons exposed to frequent hypoglycemic episodes during tight glycemic control, autonomic warning signals of hypoglycemia are less common or even absent. This dangerous acquired condition is termed “hypoglycemic unawareness.” When patients lack the early warning signs of low blood glucose, they may not take corrective measures in time. In patients with persistent, untreated hypoglycemia, the manifestations of insulin excess may develop—confusion, weakness, bizarre behavior, coma, seizures—at which point they may not be able to procure or safely swallow glucose-containing foods. Hypoglycemic awareness may be restored by preventing frequent hypoglycemic episodes. An identification bracelet, necklace, or card in the wallet or purse, as well as some form of rapidly absorbed glucose, should be carried by every diabetic person who is receiving hypoglycemic drug therapy.

2. Treatment of hypoglycemiaAll the manifestations of hypoglycemia are relieved by glucose administration. To expedite absorption, simple sugar or glucose should be given, preferably in liquid form. To treat mild hypoglycemia in a patient who is conscious and able to swallow, dextrose tablets, glucose gel, or any sugar-containing beverage or food may be given. If more severe hypoglycemia has produced unconsciousness or stupor, the treatment of choice is to give 20–50 mL of 50% glucose solution by intravenous infusion over a period of 2–3 minutes. If intravenous therapy is not available, 1 mg of glucagon injected either subcutaneously or intramuscularly may restore consciousness within 15 minutes to permit ingestion of sugar. If the patient is stuporous and glucagon is not available, small amounts of honey or syrup can be inserted into the buccal pouch. In general, however, oral feeding is contraindicated in unconscious patients. Emergency medical services should be called immediately for all episodes of severely impaired consciousness.

B. Immunopathology of Insulin Therapy

At least five molecular classes of insulin antibodies may be produced in diabetics during the course of insulin therapy: IgA, IgD, IgE, IgG, and IgM. There are two major types of immune disorders in these patients:

1. Insulin allergyInsulin allergy, an immediate type hypersensitivity, is a rare condition in which local or systemic urticaria results from histamine release from tissue mast cells sensitized by anti-insulin IgE antibodies. In severe cases, anaphylaxis results. Because sensitivity is often to noninsulin protein contaminants, the human and analog insulins have markedly reduced the incidence of insulin allergy, especially local reactions.

2. Immune insulin resistanceA low titer of circulating IgG anti-insulin antibodies that neutralize the action of insulin to a negligible extent develops in most insulin-treated patients. Rarely, the titer of insulin antibodies leads to insulin resistance and may be associated with other systemic autoimmune processes such as lupus erythematosus.

C. Lipodystrophy at Injection Sites

Injection of animal insulin preparations sometimes led to atrophy of subcutaneous fatty tissue at the site of injection. Since the development of human and analog insulin preparations of neutral pH, this type of immune complication is almost never seen. Injection of these newer preparations directly into the atrophic area often results in restoration of normal contours.

Hypertrophy of subcutaneous fatty tissue remains a problem if injected repeatedly at the same site. However, this may be corrected by avoiding the specific injection site or by liposuction.

D. Increased Cancer Risk

An increased risk of cancer attributed to insulin resistance and hyperinsulinemia has been reported in individuals with insulin resistance, prediabetes, and type 2 diabetes. Treatment with insulin and sulfonylureas, which increase circulating insulin levels, but not metformin possibly exacerbates that risk. These epidemiologic observations are preliminary and have not changed prescribing guidelines.


Several categories of oral antidiabetic agents are now available in the USA for the treatment of persons with type 2 diabetes: (1) agents that bind to the sulfonylurea receptor and stimulate insulin secretion (sulfonylureas, meglitinides, D-phenylalanine derivatives); (2) agents that lower glucose levels by their actions on liver, muscle, and adipose tissue (biguanides, thiazolidinediones); (3) agents that principally slow the intestinal absorption of glucose (α-glucosidase inhibitors); (4) agents that mimic incretin effect or prolong incretin action (glucagon-like peptide-1 [GLP-1] receptor agonists, dipeptidyl peptidase-4 [DPP-4] inhibitors), (5) agents that inhibit the reabsorption of glucose in the kidney (sodium-glucose co-transporter inhibitors [SGLTs]), and (6) agents that act by other or ill-defined mechanisms (pramlintide, bromocriptine, colesevelam).



Mechanism of Action

The major action of sulfonylureas is to increase insulin release from the pancreas (Table 41–5). They bind to a 140-kDa high-affinity sulfonylurea receptor that is associated with a beta-cell inward rectifier ATP-sensitive potassium channel (Figure 41–2). Binding of a sulfonylurea inhibits the efflux of potassium ions through the channel and results in depolarization. Depolarization opens a voltage-gated calcium channel and results in calcium influx and the release of preformed insulin.

TABLE 41–5 Regulation of insulin release in humans.


Efficacy & Safety of the Sulfonylureas

Sulfonylureas are metabolized by the liver and, with the exception of acetohexamide, the metabolites are either weakly active or inactive. The metabolites are excreted by the kidney and in the case of the second-generation sulfonylureas, partly excreted in the bile. Idiosyncratic reactions are rare, with skin rashes or hematologic toxicity (leukopenia, thrombocytopenia) occurring in less than 0.1% of cases. The second-generation sulfonylureas have greater affinity for their receptor compared with the first-generation agents. The correspondingly lower effective doses and plasma levels of the second-generation drugs therefore lower the risk of drug-drug interactions based on competition for plasma binding sites or hepatic enzyme action.

In 1970, the University Group Diabetes Program (UGDP) in the USA reported that the number of deaths due to cardiovascular disease in diabetic patients treated with tolbutamide was excessive compared with either insulin-treated patients or those receiving placebos. Owing to design flaws, this study and its conclusions were not generally accepted. In the United Kingdom, the UKPDS did not find an untoward cardiovascular effect of sulfonylurea usage in their large, long-term study. The sulfonylureas continue to be widely prescribed, and six are available in the USA (Table 41–6).

TABLE 41–6 Sulfonylureas.



Tolbutamide is well absorbed but rapidly metabolized in the liver. Its duration of effect is relatively short (6–10 hours), with an elimination half-life of 4–5 hours, and it is best administered in divided doses (eg, 500 mg before each meal). Some patients only need one or two tablets daily. The maximum dosage is 3000 mg daily. Because of its short half-life and inactivation by the liver, it is relatively safe in the elderly and in patients with renal impairment. Prolonged hypoglycemia has been reported rarely, mostly in patients receiving certain antibacterial sulfonamides (sulfisoxazole), phenylbutazone for arthralgias, or the oral azole antifungal medications to treat candidiasis. These drugs inhibit the metabolism of tolbutamide in the liver and increase its circulating levels.

Chlorpropamide has a half-life of 32 hours and is slowly metabolized in the liver to products that retain some biologic activity; approximately 20–30% is excreted unchanged in the urine. The average maintenance dosage is 250 mg daily, given as a single dose in the morning. Prolonged hypoglycemic reactions are more common in elderly patients, and the drug is contraindicated in this group. Other adverse effects include a hyperemic flush after alcohol ingestion in genetically predisposed patients and hyponatremia due to its effect on vasopressin secretion and action.

Tolazamide is comparable to chlorpropamide in potency but has a shorter duration of action. Tolazamide is more slowly absorbed than the other sulfonylureas, and its effect on blood glucose does not appear for several hours. Its half-life is about 7 hours. Tolazamide is metabolized to several compounds that retain hypoglycemic effects. If more than 500 mg/d are required, the dosage should be divided and given twice daily.

Acetohexamide is no longer available in the United States. Its half-life is only about 1 hour but its more active metabolite, hydroxyhexamide, has a half-life of 4–6 hours; thus the drug duration of action is 8–24 hours. Where available, its dosage is 0.25–1.5 g/d as single dose or in two divided doses.

Chlorpropamide, tolazamide, and acetohexamide are now rarely used in clinical practice.


Glyburide, glipizide, gliclazide, and glimepiride are 100–200 times more potent than tolbutamide. They should be used with caution in patients with cardiovascular disease or in elderly patients, in whom hypoglycemia would be especially dangerous.

Glyburide is metabolized in the liver into products with very low hypoglycemic activity. The usual starting dosage is 2.5 mg/d or less, and the average maintenance dosage is 5–10 mg/d given as a single morning dose; maintenance dosages higher than 20 mg/d are not recommended. A formulation of “micronized” glyburide (Glynase PresTab) is available in a variety of tablet sizes. However, there is some question as to its bioequivalence with nonmicronized formulations, and the FDA recommends careful monitoring to re-titrate dosage when switching from standard glyburide doses or from other sulfonylurea drugs.

Glyburide has few adverse effects other than its potential for causing hypoglycemia. Flushing has rarely been reported after ethanol ingestion, and the compound slightly enhances free water clearance. Glyburide is contraindicated in the presence of hepatic impairment and in patients with renal insufficiency.

Glipizide has the shortest half-life (2–4 hours) of the more potent agents. For maximum effect in reducing postprandial hyperglycemia, this agent should be ingested 30 minutes before breakfast because absorption is delayed when the drug is taken with food. The recommended starting dosage is 5 mg/d, with up to 15 mg/d given as a single dose. When higher daily dosages are required, they should be divided and given before meals. The maximum total daily dosage recommended by the manufacturer is 40 mg/d, although some studies indicate that the maximum therapeutic effect is achieved by 15–20 mg of the drug. An extended-release preparation (Glucotrol XL) provides 24-hour action after a once-daily morning dose (maximum of 20 mg/d). However, this formulation appears to have sacrificed its lower propensity for severe hypoglycemia compared with longer-acting glyburide without showing any demonstrable therapeutic advantages over the latter (which can be obtained as a generic drug). At least 90% of glipizide is metabolized in the liver to inactive products, and the remainder is excreted unchanged in the urine. Glipizide therapy is therefore contraindicated in patients with significant hepatic impairment. Because of its lower potency and shorter duration for action, it is preferable to glyburide in the elderly.

Glimepiride is approved for once-daily use as monotherapy or in combination with insulin. Glimepiride achieves blood glucose lowering with the lowest dosage of any sulfonylurea compound. A single daily dose of 1 mg has been shown to be effective, and the recommended maximal daily dosage is 8 mg. Glimepiride’s half-life under multidose conditions is 5–9 hours. It is completely metabolized by the liver to metabolites with weak or no activity.

Gliclazide (not available in the United States) has a half-life of 10 hours. The recommended starting dosage is 40–80 mg daily with a maximum dosage of 320 mg daily. Higher dosages are usually divided and given twice a day. It is completely metabolized by the liver to inactive metabolites.


Repaglinide is the first member of the meglitinide group of insulin secretagogues (Table 41–7). These drugs modulate beta-cell insulin release by regulating potassium efflux through the potassium channels previously discussed. There is overlap with the sulfonylureas in their molecular sites of action because the meglitinides have two binding sites in common with the sulfonylureas and one unique binding site.

TABLE 41–7 Other insulin secretagogues.


Repaglinide has a fast onset of action, with a peak concentration and peak effect within approximately 1 hour after ingestion, but the duration of action is 4–7 hours. It is cleared by hepatic CYP3A4 with a plasma half-life of 1 hour. Because of its rapid onset, repaglinide is indicated for use in controlling postprandial glucose excursions. The drug should be taken just before each meal in doses of 0.25–4 mg (maximum 16 mg/d); hypoglycemia is a risk if the meal is delayed or skipped or contains inadequate carbohydrate. It can be used in patients with renal impairment and in the elderly. Repaglinide is approved as monotherapy or in combination with biguanides. There is no sulfur in its structure, so repaglinide may be used in type 2 diabetics with sulfur or sulfonylurea allergy.

Mitiglinide (not available in the United States) is a benzylsuccinic acid derivative that binds to the sulfonylurea receptor and is similar to repaglinide in its clinical effects. It has been approved for use in Japan.


Nateglinide, a D-phenylalanine derivative, stimulates rapid and transient release of insulin from beta cells through closure of the ATP-sensitive K+ channel. It is absorbed within 20 minutes after oral administration with a time to peak concentration of less than 1 hour and is metabolized in the liver by CYP2C9 and CYP3A4 with a half-life of about 1 hour. The overall duration of action is about 4 hours. It is taken before the meal and reduces the postprandial rise in blood glucose levels. It is available as 60 and 120 mg tablets. The lower dose is used in patients with mild elevations in HbA1c. Nateglinide is efficacious when given alone or in combination with nonsecretagogue oral agents (such as metformin). Hypoglycemia is the main adverse effect. It can be used in patients with renal impairment and in the elderly.



The structure of metformin is shown below. Phenformin (an older biguanide) was discontinued in the USA because of its association with lactic acidosis. Metformin is the only biguanide currently available in the United States.


Mechanisms of Action

A full explanation of the mechanism of action of the biguanides remains elusive, but their primary effect is to activate the enzyme AMP-activated protein kinase (AMPK) and reduce hepatic glucose production. Patients with type 2 diabetes have considerably less fasting hyperglycemia as well as lower postprandial hyperglycemia after administration of biguanides; however, hypoglycemia during biguanide therapy is rare. These agents are therefore more appropriately termed “euglycemic” agents.

Metabolism & Excretion

Metformin has a half-life of 1.5–3 hours, is not bound to plasma proteins, is not metabolized, and is excreted by the kidneys as the active compound. As a consequence of metformin’s blockade of gluconeogenesis, the drug may impair the hepatic metabolism of lactic acid. In patients with renal insufficiency, biguanides accumulate and thereby increase the risk of lactic acidosis, which appears to be a dose-related complication. In the United States, metformin use is not recommended at or above a serum creatinine level of 1.4 mg/dL in women and 1.5 mg/dL in men. In the United Kingdom, it is recommended to reassess use if the serum creatinine exceeds 1.5 mg/dL (estimated glomerular filtration rate [GFR] < 45 mL/min/1.73 m2) and to stop if the serum creatinine exceeds 1.7 mg/dL (estimated GFR < 30 mL/min/1.73 m2).

Clinical Use

Biguanides are recommended as first-line therapy for type 2 diabetes. Because metformin is an insulin-sparing agent and does not increase body weight or provoke hypoglycemia, it offers obvious advantages over insulin or sulfonylureas in treating hyperglycemia in such persons. The UKPDS reported that metformin therapy decreases the risk of macrovascular as well as microvascular disease; this is in contrast to the other therapies, which only modified microvascular morbidity. Biguanides are also indicated for use in combination with insulin secretagogues or thiazolidinediones in type 2 diabetics in whom oral monotherapy is inadequate. Metformin is useful in the prevention of type 2 diabetes; the landmark Diabetes Prevention Program concluded that metformin is efficacious in preventing the new onset of type 2 diabetes in middle-aged, obese persons with impaired glucose tolerance and fasting hyperglycemia. It is interesting that metformin did not prevent diabetes in older, leaner prediabetics.

Although the recommended maximal dosage is 2.55 g daily, little benefit is seen above a total dosage of 2000 mg daily. Treatment is initiated at 500 mg with a meal and increased gradually in divided doses. Common schedules would be 500 mg once or twice daily increased to 1000 mg twice daily. The maximal dosage is 850 mg three times a day. Epidemiologic studies suggest that metformin use may reduce the risk of some cancers. These data are still preliminary, and the speculative mechanism of action is a decrease in insulin (which also functions as a growth factor) levels as well as direct cellular effects mediated by AMPK. Other studies suggest a reduction in cardiovascular deaths in humans and an increase in longevity in mice (see Chapter 60).


The most common toxic effects of metformin are gastrointestinal (anorexia, nausea, vomiting, abdominal discomfort, and diarrhea), which occur in up to 20% of patients. They are dose related, tend to occur at the onset of therapy, and are often transient. However, metformin may have to be discontinued in 3–5% of patients because of persistent diarrhea.

Metformin interferes with the calcium-dependent absorption of vitamin B12-intrinsic factor complex in the terminal ileum, and vitamin B12 deficiency can occur after many years of metformin use. Periodic screening for vitamin B12 deficiency should be considered, especially in patients with peripheral neuropathy or macrocytic anemia. Increased intake of calcium may prevent the metformin-induced B12malabsorption.

Lactic acidosis can sometimes occur with metformin therapy. It is more likely to occur in conditions of tissue hypoxia when there is increased production of lactic acid and in renal failure when there is decreased clearance of metformin. Almost all reported cases have involved patients with associated risk factors that should have contraindicated its use (kidney, liver, or cardiorespiratory insufficiency; alcoholism). Radiocontrast administration can cause acute kidney failure in patients with diabetes and incipient nephropathy. Metformin therapy should therefore be temporarily halted on the day of radiocontrast use and restarted a day or two later after confirmation that renal function has not deteriorated.


Thiazolidinediones act to decrease insulin resistance. They are ligands of peroxisome proliferator-activated receptor-gamma (PPAR-γ), part of the steroid and thyroid superfamily of nuclear receptors. These PPAR receptors are found in muscle, fat, and liver. PPAR-γ receptors modulate the expression of the genes involved in lipid and glucose metabolism, insulin signal transduction, and adipocyte and other tissue differentiation. Observed effects of the thiazolidinediones include increased glucose transporter expression (GLUT 1 and GLUT 4), decreased free fatty acid levels, decreased hepatic glucose output, increased adiponectin and decreased release of resistin from adipocytes, and increased differentiation of preadipocytes to adipocytes. Thiazolidinediones have also been shown to decrease levels of plasminogen activator inhibitor type 1, matrix metalloproteinase-9, C-reactive protein, and interleukin-6. Two thiazolidinediones are currently available: pioglitazone and rosiglitazone (Table 41–8). Their distinct side chains create differences in therapeutic action, metabolism, metabolite profile, and adverse effects. An earlier compound, troglitazone, was withdrawn from the market because of hepatic toxicity thought to be related to its side chain.

TABLE 41–8 Thiazolidinediones.


Pioglitazone has some PPAR-α as well as PPAR-γ activity. It is absorbed within 2 hours of ingestion; although food may delay uptake, total bioavailability is not affected. Absorption is decreased with concomitant use of bile acid sequestrants. Pioglitazone is metabolized by CYP2C8 and CYP3A4 to active metabolites. The bioavailability of numerous other drugs also degraded by these enzymes may be affected by pioglitazone therapy, including estrogen-containing oral contraceptives; additional methods of contraception are advised. Pioglitazone may be taken once daily; the usual starting dosage is 15–30 mg/d, and the maximum is 45 mg/d. Pioglitazone is approved as a monotherapy and in combination with metformin, sulfonylureas, and insulin for the treatment of type 2 diabetes.

Rosiglitazone is rapidly absorbed and highly protein-bound. It is metabolized in the liver to minimally active metabolites, predominantly by CYP2C8 and to a lesser extent by CYP2C9. It is administered once or twice daily; 2–8 mg is the usual total dosage. Rosiglitazone is approved for use in type 2 diabetes as monotherapy, in double combination therapy with a biguanide or sulfonylurea, or in quadruple combination with a biguanide, sulfonylurea, and insulin.

The combination of a thiazolidinedione and metformin has the advantage of not causing hypoglycemia.

These drugs also have some additional effects apart from glucose lowering. Pioglitazone lowers triglycerides and increases HDL cholesterol without affecting total cholesterol and low-density lipoprotein (LDL) cholesterol. Rosiglitazone increases total cholesterol, HDL cholesterol, and LDL cholesterol but does not have significant effect on triglycerides. These drugs have been shown to improve the biochemical and histologic features of nonalcoholic fatty liver disease. They seem to have a positive effect on endothelial function: pioglitazone reduces neointimal proliferation after coronary stent placement, and rosiglitazone has been shown to reduce microalbuminuria.

Safety concerns and troublesome side effects have significantly reduced the use of this class of drugs. A meta-analysis of 42 randomized clinical trials with rosiglitazone suggested that this drug increased the risk of angina pectoris or myocardial infarction. As a result, its use was suspended in Europe and severely restricted in the United States. A subsequent large prospective clinical trial (the RECORD study) failed to confirm the meta-analysis finding and so the United States restrictions have been lifted. The drug remains unavailable in Europe.

Fluid retention occurs in about 3–4 % patients on thiazolidinedione monotherapy and occurs more frequently (10–15%) in patients on concomitant insulin therapy. Heart failure can occur, and the drugs are contraindicated in patients with New York Heart Association class III and IV cardiac status (see Chapter 13). Macular edema is a rare side effect that improves when the drug is discontinued. Loss of bone mineral density and increased atypical extremity bone fractures in women are described for both compounds, which is postulated to be due to decreased osteoblast formation. Other side effects include anemia, which might be due to a dilutional effect of increased plasma volume rather than a reduction in red cell mass. Weight gain occurs, especially when used in combination with a sulfonylurea or insulin. Some of the weight gain is fluid retention but there is also an increase in total fat mass. In preclinical trials, bladder tumors were observed in male rats on pioglitazone. A planned interim analysis of a long-term observational cohort of patients treated with pioglitazone found an increased risk of bladder cancer with increased dosage and duration of drug use. Safety analysis of a study designed to evaluate the impact of pioglitazone on macrovascular events noted 14 cases of bladder cancer in the treated group and 5 cases in the placebo group, a significant difference. Although there are currently no recommendations regarding screening for bladder cancer, it should be considered in patients on long-term therapy.

Troglitazone, the first medication in this class, was withdrawn because of cases of fatal liver failure. Although rosiglitazone and pioglitazone have not been reported to cause liver injury, the drugs are not recommended for use in patients with active liver disease or pretreatment elevation of alanine aminotransferase (ALT) 2.5 times greater than normal. Liver function tests should be performed prior to initiation of treatment and periodically thereafter.


The α-glucosidase inhibitors competitively inhibit the intestinal α-glucosidase enzymes and reduce postmeal glucose excursions by delaying the digestion and absorption of starch and disaccharides (Table 41–9). Acarbose and miglitol are available in the United States. Voglibose is available in Japan, Korea, and India. Acarbose and miglitol are potent inhibitors of glucoamylase, α-amylase, and sucrase but have less effect on isomaltase and hardly any on trehalase and lactase. Acarbose has the molecular mass and structural features of a tetrasaccharide and very little is absorbed. In contrast, miglitol has structural similarity to glucose and is absorbed.

TABLE 41–9 Alpha-glucosidase inhibitors.


Acarbose treatment is initiated at a dosage of 50 mg twice daily with gradual increase to 100 mg three times a day. It lowers postprandial glucose levels by 30–50%. Miglitol therapy is initiated at a dosage of 25 mg three times a day. The usual maintenance dosage is 50 mg three times a day but some patients may need 100 mg three times a day. The drug is not metabolized and is cleared by the kidney. It should not be used in renal failure.

Prominent adverse effects of α-glucosidase inhibitors include flatulence, diarrhea, and abdominal pain and result from the appearance of undigested carbohydrate in the colon that is then fermented into short-chain fatty acids, releasing gas. These adverse effects tend to diminish with ongoing use because chronic exposure to carbohydrate induces the expression of α-glucosidase in the jejunum and ileum, increasing distal small intestine glucose absorption and minimizing the passage of carbohydrate into the colon. Although not a problem with monotherapy or combination therapy with a biguanide, hypoglycemia may occur with concurrent sulfonylurea treatment. Hypoglycemia should be treated with glucose (dextrose) and not sucrose, whose breakdown may be blocked. An increase in hepatic aminotransferases has been noted in clinical trials with acarbose, especially with dosages greater than 300 mg/d. The abnormalities resolve on stopping the drug.

These drugs are infrequently prescribed in the United States because of their prominent gastrointestinal adverse effects and relatively modest glucose-lowering benefit.


An oral glucose load provokes a higher insulin response compared with an equivalent dose of glucose given intravenously. This is because the oral glucose causes a release of gut hormones (“incretins”), principally glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic peptide (GIP), that amplify the glucose-induced insulin secretion. When GLP-1 is infused in patients with type 2 diabetes, it stimulates insulin release and lowers glucose levels. The GLP-1 effect is glucose dependent in that the insulin release is more pronounced when glucose levels are elevated but less so when glucose levels are normal. For this reason, GLP-1 has a lower risk for hypoglycemia than the sulfonylureas. In addition to its insulin stimulatory effect, GLP-1 has a number of other biologic effects. It suppresses glucagon secretion, delays gastric emptying, and reduces apoptosis of human islets in culture. In animals, GLP-1 inhibits feeding by a central nervous system mechanism. Type 2 diabetes patients on GLP-1 therapy are less hungry. It is unclear whether this is mainly related to the deceleration of gastric emptying or whether there is a central nervous system effect as well.

GLP-1 is rapidly degraded by dipeptidyl peptidase-4 (DPP-4) and by other enzymes such as endopeptidase 24.11 and is also cleared by the kidney. The native peptide therefore cannot be used therapeutically. One approach to this problem has been to develop metabolically stable analogs or derivatives of GLP-1 that are not subject to the same enzymatic degradation or renal clearance. Four such GLP-1 receptor agonists, exenatide, liraglutide, albiglutide, and dulaglutide are available for clinical use. The other approach has been to develop inhibitors of DPP-4 and prolong the action of endogenously released GLP-1 and GIP. Four oral DPP-4 inhibitors, sitagliptin, saxagliptin, linagliptin, and alogliptin, are available in the United States. An additional inhibitor, vildagliptin, is available in Europe.


Exenatide, a derivative of the exendin-4 peptide in Gila monster venom, has a 53% homology with native GLP-1, and a glycine substitution to reduce degradation by DPP-4. Exenatide is approved as an injectable, adjunctive therapy in persons with type 2 diabetes treated with metformin or metformin plus sulfonylureas who still have suboptimal glycemic control.

Exenatide is dispensed as fixed-dose pens (5 mcg and 10 mcg). It is injected subcutaneously within 60 minutes before breakfast and dinner. It reaches a peak concentration in approximately 2 hours with a duration of action of up to 10 hours. Therapy is initiated at 5 mcg twice daily for the first month and if tolerated can be increased to 10 mcg twice daily. Exenatide LAR is a once-weekly preparation that is dispensed as a powder (2 mg). It is suspended in the provided diluent just prior to injection. When exenatide is added to preexisting sulfonylurea therapy, the oral hypoglycemic dosage may need to be decreased to prevent hypoglycemia. The major adverse effect is nausea (about 44% of users), which is dose dependent and declines with time. Exenatide mono-and combination therapy results in HbA1creductions of 0.2–1.2%. Weight loss in the range of 2–3 kg occurs and contributes to the improvement of glucose control. In comparative trials the long-acting (LAR) preparation lowers the HbA1c level a little more than the twice-daily preparation. Exenatide undergoes glomerular filtration, and the drug is not approved for use in patients with estimated GFR of less than 30 mL/min.

High-titer antibodies against exenatide develop in about 6% of patients and in half of these patients an attenuation of glycemic response has been seen.

Liraglutide is a soluble fatty acid-acylated GLP-1 analog. The half-life is approximately 12 hours permitting once-daily dosing. It is approved in patients with type 2 diabetes who achieve inadequate control with diet and exercise and are receiving concurrent treatment with metformin, sulfonylureas, or thiazolidinediones. Treatment is initiated at 0.6 mg and increased after 1 week to 1.2 mg daily. If needed the dosage can be increased to 1.8 mg daily. In clinical trials liraglutide results in a reduction of HbA1c of 0.8–1.5%; weight loss ranges from none to 3.2 kg. The most frequent side effects are nausea (28%) and vomiting (10%).

Albiglutide is a human GLP-1 dimer fused to human albumin. The half-life of albiglutide is about 5 days and a steady state is reached after 4–5 weeks of once weekly administration. The usual dose is 30 mg weekly by subcutaneous injection. The drug is supplied in a self-injection pen containing a powder that is reconstituted just prior to administration. Weight loss is much less common than with exenatide and liraglutide. The most frequent adverse effects were nausea and injection site erythema.

Dulaglutide consists of two GLP-1 analog molecules covalently linked to an Fc fragment of human IgG4. The GLP-1 molecule has amino acid substitutions that resist DPP-4 action. The half-life of dulaglutide is about 5 days. The usual dose is 0.75 mg weekly by subcutaneous injection. The maximum recommended dose is 1.5 mg weekly. The most frequent adverse reactions were nausea, diarrhea and vomiting.

All of the GLP-1 receptor agonists may increase the risk of pancreatitis. Patients on these drugs should be counseled to seek immediate medical care if they experience unexplained persistent severe abdominal pain. Cases of renal impairment and acute renal injury have been reported in patients taking exenatide. Some of these patients had preexisting kidney disease or other risk factors for renal injury. A number of them reported having nausea, vomiting and diarrhea and it is possible that volume depletion contributed to the development of renal injury. Both exenatide and liraglutide stimulate thyroidal C-cell (parafollicular) tumors in rodents. Human thyroidal C-cells express very few GLP-1 receptors and the relevance to human therapy is unclear. The drugs, however, should not be used in persons with a past medical or family history of medullary thyroid cancer or multiple endocrine neoplasia (MEN) syndrome type 2.


Sitagliptin is given orally as 100 mg once daily, has an oral bioavailability of over 85%, achieves peak concentrations within 1–4 hours, and has a half-life of approximately 12 hours. It is primarily excreted in the urine, in part by active tubular secretion of the drug. Hepatic metabolism is limited and mediated largely by the cytochrome CYP3A4 isoform and, to a lesser degree, by CYP2C8. The metabolites have insignificant activity. Dosage should be reduced in patients with impaired renal function (50 mg if estimated GFR is 30–50 mL/min and 25 mg if < 30 mL/min). Sitagliptin has been studied as monotherapy and in combination with metformin, sulfonylureas, and thiazolidinediones. Therapy with sitagliptin has resulted in HbA1c reductions of 0.5–1.0%.

Common adverse effects include nasopharyngitis, upper respiratory infections, and headaches. Hypoglycemia can occur when the drug is combined with insulin secretagogues or insulin. There have been postmarketing reports of acute pancreatitis (fatal and nonfatal) and severe allergic and hypersensitivity reactions. Sitagliptin should be immediately discontinued if pancreatitis or allergic and hypersensitivity reactions occur.

Saxagliptin is given orally as 2.5–5 mg daily. The drug reaches maximal concentrations within 2 hours (4 hours for its active metabolite). It is minimally protein bound and undergoes hepatic metabolism by CYP3A4/5. The major metabolite is active, and excretion is by both renal and hepatic pathways. The terminal plasma half-life is 2.5 hours for saxagliptin and 3.1 hours for its active metabolite. Dosage adjustment is recommended for individuals with renal impairment and concurrent use of strong CYP3A4/5 inhibitors such as antiviral, antifungal, and certain antibacterial agents. It is approved as monotherapy and in combination with biguanides, sulfonylureas, and thiazolidinediones. During clinical trials, mono- and combination therapy with saxagliptin resulted in an HbA1c reduction of 0.4–0.9%.

Adverse effects include an increased rate of infections (upper respiratory tract and urinary tract), headaches, and hypersensitivity reactions (urticaria, facial edema). The dosage of a concurrently administered insulin secretagogue or insulin may need to be lowered to prevent hypoglycemia.

Linagliptin lowers HbA1c by 0.4–0.6% when added to metformin, sulfonylurea, or pioglitazone. The dosage is 5 mg daily and since it is primarily excreted via the bile, no dosage adjustment is needed in renal failure.

Adverse reactions include nasopharyngitis and hypersensitivity reactions (urticaria, angioedema, localized skin exfoliation, bronchial hyperreactivity). The risk of pancreatitis may be increased.

Vildagliptin (not available in the United States) lowers HbA1c levels by 0.5–1% when added to the therapeutic regimen of patients with type 2 diabetes. The dosage is 50 mg once or twice daily. Adverse reactions include upper respiratory infections, nasopharyngitis, dizziness, and headache. Rarely, it can cause hepatitis and liver function tests should be performed quarterly in the first year of use and periodically thereafter.

In animal studies, high doses of DPP-4 inhibitors and GLP-1 agonists cause expansion of pancreatic ductal glands and generation of premalignant pancreatic intraepithelial (PanIN) lesions that have the potential to progress to pancreatic adenocarcinoma. The relevance to human therapy is unclear and currently there is no evidence that these drugs cause pancreatic cancer in humans.


Glucose is freely filtered by the renal glomeruli and is reabsorbed in the proximal tubules by the action of sodium-glucose transporters (SGLTs). Sodium-glucose transporter 2 (SGLT2) accounts for 90% of glucose reabsorption, and its inhibition causes glycosuria and lowers glucose levels in patients with type 2 diabetes. The SGLT2 inhibitors canagliflozin, dapagliflozin, and empagliflozin are approved for clinical use.

Canagliflozin reduces the threshold for glycosuria from a plasma glucose threshold of approximately 180 mg/dL to 70–90 mg/dL. It has been shown to reduce HbA1c by 0.6–1% when used alone or in combination with other oral agents or insulin. It also results in modest weight loss of 2–5 kg. The usual dosage is 100 mg daily. Increasing the dosage to 300 mg daily in patients with normal renal function can lower the HbA1c by an additional 0.5%.

Dapagliflozin reduces HbA1c by 0.5–0.8% when used alone or in combination with other oral agents or insulin. It also results in modest weight loss of about 2–4 kg. The usual dosage is 10 mg daily but 5 mg daily is recommended initially in patients with hepatic failure.

Empagliflozin reduces HbA1c by 0.5–0.7% when used alone or in combination with other oral agents or insulin. It also results in modest weight loss of 2–3 kg. The usual dosage is 10 mg daily but 25 mg/d may be used.

As might be expected, the efficacy of the SGLT2 inhibitors is reduced in chronic kidney disease. Canagliflozin and empagliflozin are contraindicated in patients with estimated GFR less than 45 mL/min/1.73 m2. Dapagliflozin is not recommended for use in patients with estimated GFR less than 60 mL/min/1.73 m2. The main side effects are increased incidence of genital infections and urinary tract infections affecting about 8–9% of patients. The osmotic diuresis can also cause intravascular volume contraction and hypotension. Canagliflozin and empagliflozin caused a modest increase in LDL cholesterol levels (4–8%). In clinical trials patients taking dapagliflozin had higher rates of breast cancer (nine cases versus none in comparator arms) and bladder cancer (nine cases versus one in placebo arm). These cancer rates exceeded the expected rates in an age-matched reference diabetes population.


Pramlintide is an islet amyloid polypeptide (IAPP, amylin) analog. IAPP is a 37-amino-acid peptide present in insulin secretory granules and secreted with insulin. It has approximately 46% homology with the calcitonin gene-related peptide (CGRP; see Chapter 17) and physiologically acts as a negative feedback on insulin secretion. At pharmacologic doses, IAPP reduces glucagon secretion, slows gastric emptying by a vagally medicated mechanism, and centrally decreases appetite. Pramlintide is an IAPP analog with substitutions of proline at positions 25, 28, and 29. These modifications make pramlintide soluble and non–self-aggregating, and suitable for pharmacologic use. Pramlintide is approved for use in insulin-treated type 1 and type 2 patients who are unable to achieve their target postprandial blood glucose levels. It is rapidly absorbed after subcutaneous administration; levels peak within 20 minutes, and the duration of action is not more than 150 minutes. It is metabolized and excreted by the kidney, but even at low creatinine clearance there is no significant change in bioavailability. It has not been evaluated in dialysis patients.

Pramlintide is injected immediately before eating; dosages range from 15 to 60 mcg subcutaneously for type 1 patients and from 60 to 120 mcg for type 2 patients. Therapy with this agent should be initiated at the lowest dosage and titrated upward. Because of the risk of hypoglycemia, concurrent rapid- or short-acting mealtime insulin dosages should be decreased by 50% or more. Pramlintide should always be injected by itself using a separate syringe; it cannot be mixed with insulin. The major adverse effects of pramlintide are hypoglycemia and gastrointestinal symptoms, including nausea, vomiting, and anorexia. Since the drug slows gastric emptying, recovery from hypoglycemia can be problematic because of the delay in absorption of fast-acting carbohydrates.

Selected patients with type 1 diabetes who have problems with postprandial hyperglycemia can use pramlintide effectively to control the glucose rise especially in the setting of a high-carbohydrate meal. The drug is not that useful in type 2 patients who can instead use the GLP-1 receptor agonists.

Colesevelam hydrochloride, the bile acid sequestrant and cholesterol-lowering drug, is approved as an antihyperglycemic therapy for persons with type 2 diabetes who are taking other medications or have not achieved adequate control with diet and exercise. The exact mechanism of action is unknown but presumed to involve an interruption of the enterohepatic circulation and a decrease in farnesoid X receptor (FXR) activation. FXR is a nuclear receptor with multiple effects on cholesterol, glucose, and bile acid metabolism. Bile acids are natural ligands of the FXR. Additionally, the drug may impair glucose absorption. In clinical trials, it lowered the HbA1c concentration 0.3–0.5%. Adverse effects include gastrointestinal complaints (constipation, indigestion, flatulence). It can also exacerbate the hypertriglyceridemia that commonly occurs in people with type 2 diabetes.

Bromocriptine, the dopamine agonist, in randomized placebo-controlled studies lowered HbA1c by 0–0.2% compared with baseline and by 0.4–0.5% compared with placebo. The mechanism by which it lowers glucose levels is not known. The main adverse events are nausea, fatigue, dizziness, vomiting, and headache.

Colesevelam and bromocriptine have very modest efficacy in lowering glucose levels and their use for this purpose is questionable.


Failure to maintain a good response to therapy over the long term owing to a progressive decrease in beta-cell mass, reduction in physical activity, decline in lean body mass, or increase in ectopic fat deposition remains a disconcerting problem in the management of type 2 diabetes. Multiple medications may be required to achieve glycemic control (Figure 41–6). Unless there is a contraindication, medical therapy should be initiated with metformin. If clinical failure occurs with metformin monotherapy, a second agent is added. Options include sulfonylureas, repaglinide or nateglinide, pioglitazone, GLP-1 receptor agonists, DPP-4 inhibitors, SGLT2 inhibitors, and insulin. In the choice of the second agent consideration should be given to efficacy of the agent, hypoglycemic risk, effect on weight, side effects, and cost. In patients who experience hyperglycemia after a carbohydrate-rich meal (such as dinner), a short-acting secretagogue before that meal may suffice to control the glucose levels. Patients with severe insulin resistance may be candidates for pioglitazone. Patients who are very concerned about weight gain may benefit from a trial of a GLP-1 receptor agonist, a DPP-4 inhibitor, or an SGLT2 inhibitor. If two agents are inadequate a third agent is added, although data regarding efficacy of such combined therapy are limited.


FIGURE 41–6 Suggested algorithm for the treatment of type 2 diabetes. The seven main classes of agents are metformin, sulfonylureas (includes nateglinide, repaglinide), pioglitazone, GLP-1 receptor agonists, DPP-4 inhibitors, SGLT2 inhibitors, insulins. (α-Glucosidase inhibitors, colesevelam, pramlintide, and bromocriptine not included because of limited efficacy and significant adverse reactions). (Data from the consensus panel of the American Diabetes Association/European Association for the Study of Diabetes, as described in Inzucchi SE et al: Diabetes Care 2012;35:1364.)

When the combination of oral agents and injectable GLP-1 receptor agonists fails to adequately control glucose levels, insulin therapy should be instituted. Various insulin regimens may be effective. Simply adding nighttime intermediate or long-acting insulin to the oral regimen may lead to improved fasting glucose levels and adequate control during the day. If daytime glucose levels are problematic, premixed insulins before breakfast and dinner may help. If such a regimen does not achieve adequate control or leads to unacceptable rates of hypoglycemia, a more intensive basal bolus insulin regimen (long-acting basal insulin) combined with rapid acting analog before meals can be instituted. Metformin has been shown to be effective when combined with insulin therapy and should be continued. Pioglitazone can be used with insulin, but this combination is associated with more weight gain and peripheral edema. Continuing with sulfonylureas, GLP-1 receptor agonists, DPP-4 inhibitors, and SGLT2 inhibitors can be of benefit in selected patients. Cost, complexity, and risk for adverse events should be considered when deciding which drugs to continue once the patient starts on insulin therapy.


Chemistry & Metabolism

Glucagon is synthesized in the alpha cells of the pancreatic islets of Langerhans (Table 41–1). Glucagon is a peptide—identical in all mammals—consisting of a single chain of 29 amino acids, with a molecular weight of 3485. Selective proteolytic cleavage converts a large precursor molecule of approximately 18,000 MW to glucagon. One of the precursor intermediates consists of a 69-amino-acid peptide called glicentin, which contains the glucagon sequence interposed between peptide extensions.

Glucagon is extensively degraded in the liver and kidney as well as in plasma and at its tissue receptor sites. Because of its rapid inactivation by plasma, chilling of the collecting tubes and addition of inhibitors of proteolytic enzymes are necessary when samples of blood are collected for immunoassay of circulating glucagon. Its half-life in plasma is between 3 and 6 minutes, which is similar to that of insulin.

Pharmacologic Effects of Glucagon

A. Metabolic Effects

The first six amino acids at the amino terminal of the glucagon molecule bind to specific Gs protein-coupled receptors on liver cells. This leads to an increase in cAMP, which facilitates catabolism of stored glycogen and increases gluconeogenesis and ketogenesis. The immediate pharmacologic result of glucagon infusion is to raise blood glucose at the expense of stored hepatic glycogen. There is no effect on skeletal muscle glycogen, presumably because of the lack of glucagon receptors on skeletal muscle. Pharmacologic amounts of glucagon cause release of insulin from normal pancreatic beta cells, catecholamines from pheochromocytoma, and calcitonin from medullary carcinoma cells.

B. Cardiac Effects

Glucagon has a potent inotropic and chronotropic effect on the heart, mediated by the cAMP mechanism described above. Thus, it produces an effect very similar to that of β-adrenoceptor agonists without requiring functioning β receptors.

C. Effects on Smooth Muscle

Large doses of glucagon produce profound relaxation of the intestine. In contrast to the above effects of the peptide, this action on the intestine may be due to mechanisms other than adenylyl cyclase activation.

Clinical Uses

A. Severe Hypoglycemia

The major use of glucagon is for emergency treatment of severe hypoglycemic reactions in patients with type 1 diabetes when unconsciousness precludes oral feedings and intravenous glucose treatment is not possible. Recombinant glucagon is currently available in 1-mg vials for parenteral (IV, IM, or SC) use (Glucagon Emergency Kit). Nasal sprays have been developed for this purpose but have not yet received FDA approval.

B. Endocrine Diagnosis

Several tests use glucagon to diagnose endocrine disorders. In patients with type 1 diabetes mellitus, a classic research test of pancreatic beta-cell secretory reserve uses 1 mg of glucagon administered as an intravenous bolus. Because insulin-treated patients develop circulating anti-insulin antibodies that interfere with radioimmunoassays of insulin, measurements of C-peptide are used to indicate beta-cell secretion.

C. Beta-Adrenoceptor Blocker Overdose

Glucagon is sometimes useful for reversing the cardiac effects of an overdose of β-blocking agents because of its ability to increase cAMP production in the heart. However, it is not clinically useful in the treatment of cardiac failure.

D. Radiology of the Bowel

Glucagon has been used extensively in radiology as an aid to X-ray visualization of the bowel because of its ability to relax the intestine.

Adverse Reactions

Transient nausea and occasional vomiting can result from glucagon administration. These are generally mild, and glucagon is relatively free of severe adverse reactions. It should not be used in a patient with pheochromocytoma.

SUMMARY Drugs Used for Diabetes







Action to Control Cardiovascular Risks in Diabetes Study Group: Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med 2008;358:2545.

Adler AI et al: Association of systolic blood pressure with macrovascular and microvascular complications of type 2 diabetes (UKPDS 36): Prospective observational study. Br Med J 2000;321:412.

ADVANCE Collaborative Group: Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med 2008;358:2560.

Ahmadian M et al: PPARγ signaling and metabolism: The good, the bad and the future. Nat Med 2013;19:557.

American Diabetes Association: Diagnosis and classification of diabetes mellitus. Diabetes Care 2013;36(Suppl 1):S67.

Andrianesis V, Doupis J: The role of kidney in glucose homeostasis—SGLT2 inhibitors, a new approach in diabetes treatment. Expert Rev Clin Pharmacol 2013;6:519.

Bennett WL et al: Comparative effectiveness and safety of medications for type 2 diabetes: An update including new drugs and 2-drug combinations. Ann Intern Med 2011;154: 602.  Erratum in: Ann Intern Med 2011;155:67.

Butler PC et al: A critical analysis of the clinical use of incretin-based therapies: Are the GLP-1 therapies safe? Diabetes Care 2013;36:2118.

Diabetes Prevention Program Research Group: Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393.

Expert Committee on the Diagnosis and Classification of Diabetes Mellitus: Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 2003;26(Suppl 1):S5.

Gaede P et al: Effect of a multifactorial intervention on mortality in type 2 diabetes. N Engl J Med 2008;358:580.

Guo S: Insulin signaling, resistance, and the metabolic syndrome: Insights from mouse models to disease mechanisms. J Endocrinol 2014;220:T1.

Inzucchi SE et al: Management of hyperglycemia in type 2 diabetes: A patient-centered approach: Position statement of the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care 2012;35:1364.

Karagiannis T et al: Dipeptidyl peptidase-4 inhibitors for treatment of type 2 diabetes mellitus in the clinical setting: Systematic review and meta-analysis. BMJ 201212;344.

Kitabchi A et al: Thirty years of personal experience in hyperglycemic crises: Diabetic ketoacidosis and hyperglycemic hyperosmolar state. J Clin Endocrinol Metab 2008;93:1541.

Lefebvre P et al: Role of bile acids and bile acid receptors in metabolic regulation. Physiol Rev 2009;89:147.

Miyazaki Y, DeFronzo RA: Rosiglitazone and pioglitazone similarly improve insulin sensitivity and secretion, glucose tolerance and adipocytokines in type 2 diabetic patients. Diabetes Obes Metab 2008;10:1204.

Nauck MA: A critical analysis of the clinical use of incretin-based therapies: The benefits by far outweigh the potential risks. Diabetes Care 2013;36:2126.

Nwose OM, Jones MR: Atypical mechanism of glucose modulation by colesevelam in patients with type 2 diabetes. Clin Med Insights Endocrinol Diabetes 2013;6:75.

Ratner RE et al: Amylin replacement with pramlintide as an adjunct to insulin therapy improves long term glycemic and weight control in type 1 diabetes mellitus: A 1-year randomized controlled trial. Diabetic Med 2004;21:1204.

Reitman ML et al: Pharmacogenetics of metformin response: A step in the path toward personalized medicine. J Clin Invest 2007;117:1226.

Rizzo M et al: Non-glycemic effects of pioglitazone and incretin-based therapies. Expert Opin Ther Targets 2013;17:739.

Russell S: Incretin-based therapies for type 2 diabetes mellitus: A review of direct comparisons of efficacy, safety and patient satisfaction. Int J Clin Pharm 2013;35:159.

Standl E, Schnell O: Alpha-glucosidase inhibitors 2012—cardiovascular considerations and trial evaluation. Diab Vasc Dis Res 2012;9:163.

Switzer SM et al: Intensive insulin therapy in patients with type 1 diabetes mellitus. Endocrinol Metab Clin North Am 2012;41:89.

Turner RM et al: Thiazolidinediones and associated risk of bladder cancer: A systematic review and meta-analysis. Br J Clin Pharmacol 2013 Dec 10.

United Kingdom Prospective Diabetes Study (UKPDS) Group: Glycemic control with diet, sulfonylurea, metformin, or insulin in patients with type 2 diabetes mellitus: Progressive requirement for multiple therapies: UKPDS 49. JAMA 1999;281:2005.

United Kingdom Prospective Diabetes Study (UKPDS) Group: Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. Br Med J 1998;317:703.


This patient has multiple risk factors for type 2 diabetes. Although she does not have a prior history of fasting hyperglycemia, glucose intolerance, or gestational diabetes, other risk factors are present. Further evaluations that should be obtained include HbA1c concentration, dilated retinal examination, baseline laboratory tests, spot urine test for microalbumin/creatinine ratio, plasma creatinine level, and neurologic examination. The patient should be taught how to use a glucose meter and monitor her fingerstick blood glucose level, referred to a nutritionist for dietary instruction, and given diabetes self-management education. Assuming she has no renal or hepatic impairment, hygienic interventions (diet and exercise) and metformin would be the first line of treatment. If she is unable to achieve adequate glycemic control on metformin, an additional agent such as an insulin secretagogue (ie, a sulfonylurea, meglitinide, or nateglinide), insulin, or another antidiabetic medication could be added.