Chapter 43: Endocrine Pancreas and Pharmacotherapy of Diabetes Mellitus and Hypoglycemia

Regulation of Blood Glucose. In healthy humans, blood glucose is tightly maintained despite wide fluctuations in glucose consumption, utilization, and production. The maintenance of glucose homeostasis, generally termed glucose tolerance, is a highly developed systemic process involving the integration of several major organs through multilayered communication (Figure 43–1). Although endocrine control of blood glucose, primarily through the actions of insulin, is of central importance, myriad levels of inter-organ communication, via other hormones, nerves, local factors and substrates, also play a vital role. The pancreatic β cell is central in this homeostatic process, adjusting the amount of insulin secreted very precisely to promote glucose uptake after meals and to regulate glucose output from the liver during fasting.

Figure 43–1.

Insulin, glucagon, and glucose homeostasis. A. Fasting State–In healthy humans plasma glucose is maintained in a range from 4.4-5 mM, and fatty acids near 400 μM. In the absence of nutrient absorption from the GI tract, glucose is supplied primarily from the liver and fatty acids from adipose tissue. With fasting, plasma insulin levels are low, and plasma glucagon is elevated, contributing to increased hepatic glycogenolysis and gluconeogenesis; low insulin also releases adipocytes from inhibition, permitting increased lipogenesis. Most tissues oxidize primarily fatty acids during fasting, sparing glucose for use by the CNS. B. Prandial State–During feeding, nutrient absorption causes an increases in plasma glucose, resulting in release of incretins from the gut and neural stimuli that promote insulin secretion. Under the control of insulin, the liver, sekletal muscle and adipose tissue actively take up glucose. Hepatic glucose production and lipolysis are inhibited, and total body glucose oxidation increases. The brain senses plasma glucose concentrations and provides regulatory inputs contributing to fuel homeostasis. The boldness of the arrows reflects relative intensity of action; a dashed line indicates little or no activity.

In the fasting state (Figure 43–1A), most of the fuel demands of the body are met by the oxidation of fatty acids. Importantly, the brain does not effectively utilize fatty acids to meet energy needs and in the fasting state requires glucose for normal function; glucose requirements are ~2 mg/kg per minute in adult humans, largely to supply the central nervous system (CNS) with an energy source. Fasting glucose requirements are primarily provided by the liver, with a minor contribution from the kidney. Liver glycogen stores provide some of this glucose; conversion of gluconeogenic precursors, primarily lactate, alanine and glycerol, into glucose accounts for the remainder. The dominant regulation of hepatic glycogenolysis and gluconeogenesis are the pancreatic islet hormones insulin and glucagon. Insulin inhibits hepatic glucose production at several levels, and the decline of circulating insulin concentrations in the postabsorptive state (fasting) is permissive for higher rates of glucose output. Glucagon maintains blood glucose concentrations at physiological levels in the absence of exogenous carbohydrate (overnight and in between meals) by stimulating gluconeogenesis and glycogenolysis by the liver.

Ingestion of a meal provides a substantial challenge to glucose homeostasis (Figure 43–1B). Adults typically consume 30-90 g of carbohydrate in a single meal, considerably more than the extracellular glucose pool, which typically contains 15-20 g of glucose. Therefore, uptake and disposition of meal-derived glucose requires effective coordination of a range of processes, from digestion and absorption to intracellular glycolysis and glycogen synthesis, to prevent major shifts in plasma glucose concentrations. Regulation of nutrient disposition after meals is under the primary control of insulin. Insulin secretion is stimulated by food ingestion, nutrient absorption, and increases in blood glucose, and insulin promotes glucose, lipid, and protein anabolism. The centrality of insulin in glucose metabolism is emphasized by the fact that all the forms of human diabetes have as a root cause some abnormality of insulin secretion or action.

In healthy humans, β cell function is primarily controlled by plasma glucose concentrations. Elevations of blood glucose are necessary for insulin release above basal levels, and other stimuli are relatively ineffective when plasma glucose is in the fasting range (4.4-5.5 mM or 80-100 mg %). These other stimuli include nutrient substrates, insulinotropic hormones released from the gastrointestinal (GI) tract, and autonomic neural pathways. As a result, the pancreatic β cell is stimulated from the time of food presentation, through nutrient absorption, and until blood glucose returns to fasting levels. Neural stimuli cause some increase of insulin secretion prior to food consumption. In addition to these cephalic responses, neural stimulation of insulin secretion occurs throughout the meal and contributes significantly to glucose tolerance. Arrival of nutrient chyme to the intestine leads to the release of insulinotropic peptides from specialized endocrine cells in the intestinal mucosa. Glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide 1 (GLP-1), together termed incretins, are the essential gut hormones contributing to glucose tolerance. They are secreted in proportion to the nutrient load ingested and relay this information to the islet as part of a feed-forward mechanism that allows an insulin response appropriate to meal size. Insulin secretion rates in healthy humans are highest in the early digestive phase of meals, preceding and limiting the peak in blood glucose. This pattern of premonitory insulin secretion is an essential feature of normal glucose tolerance. How to mimic this pattern is one of the key challenges for successful insulin therapy in diabetic patients.

Elevated circulating insulin concentrations lower glucose in blood by inhibiting hepatic glucose production and stimulating the uptake and metabolism of glucose by muscle and adipose tissue. These two important effects occur at different concentrations of insulin. Production of glucose is inhibited half-maximally by an insulin concentration of ~120 pmol/L, whereas glucose utilization is stimulated half-maximally at ~300 pmol/L. Some of the effects of insulin on the liver occur rapidly, within the first 20 minutes of meal ingestion, whereas stimulation of peripheral glucose uptake may require up to an hour to reach significant rates. This is probably due to the quick access of insulin to hepatocytes, via the hepatic-portal circulation and the liver sinusoids, and the slower passage of insulin to its receptors on muscle and adipose cells. Insulin has potent effects to reduce lipolysis from adipocytes, primarily through the inhibition of hormone-sensitive lipase, and increases lipid storage by promoting lipoprotein lipase synthesis and adipocyte glucose uptake. Finally, insulin stimulates amino acid uptake and protein synthesis and inhibits protein degradation in muscle and other tissues; it thus causes a decrease in the circulating concentrations of most amino acids.

The demand for glucose as an energy source for skeletal muscle increases dramatically during exercise. Glycogen stored in skeletal muscle is mobilized for some of these needs, but there are limited supplies that are used, mostly at the onset of activity. Most of the glucose support of exercise comes from hepatic gluconeogenesis. The dominant regulation of hepatic glucose production during exercise comes from epinephrine and norepinephrine. The catecholamines stimulate glycogenolysis and gluconeogenesis, inhibit insulin secretion, and enhance release of glucagon, all contributing to increased hepatic glucose output. In addition, catecholamines promote lipolysis, freeing fatty acids for oxidation in exercising muscle and glycerol for hepatic gluconeogenesis.

Pancreatic Islet Physiology and Insulin Secretion. The pancreatic islets comprise 1-2% of the pancreatic volume and are scattered throughout the exocrine pancreas. The pancreatic islet is a highly vascularized, highly innervated mini-organ containing five endocrine cell types: α cells that secrete glucagon, β cells that secrete glucose, δ cells that secrete somatostatin, cells that secrete pancreatic polypeptide, and ∊ cells that secrete ghrelin. The endocrine, exocrine, and ductal cells of the pancreas share a common embryologic heritage but in the adult are functionally distinct. Insulin and glucagon are important pharmacological agents in the treatment of diabetes, and an understanding of their synthesis, secretion, and action are important in the therapeutic efforts related to diabetes.

Insulin is initially synthesized as a single polypeptide chain, preproinsulin (110 amino acid), which is processed first to proinsulin and then to insulin and C-peptide (Figure 43–2). This complex and highly regulated process involves the Golgi complex, the endoplasmic reticulum, and importantly the distinctive secretory granules of the β cell. Secretory granules are critical not only in bringing insulin to the cell surface for exocytosis, but also in the cleavage and processing of the prohormone to the final secretion products, insulin and C-peptide. The structure of insulin is discussed in the section describing its therapeutic use. Equimolar quantities of insulin and C-peptide (31 amino acids) are co-secreted. Insulin has a t_1/2 of 5-6 minutes due to extensive hepatic clearance. C-peptide, in contrast, with no known physiological function or receptor, has a t_1/2 of ~30 minutes. Because almost all of the C-peptide released into the portal vein reaches the peripheral circulation where it can be measured, this peptide is useful in assessment of β-cell secretion, and to distinguish endogenous and exogenous hyperinsulinemia (for example in the evaluation of insulin-induced hypoglycemia). Importantly, the β cell also synthesizes and secretes islet amyloid polypeptide (IAPP) or amylin, a 37–amino acid peptide. IAPP is not essential for life but influences GI motility and the speed of glucose absorption. Pramlintide is an agent used in the treatment of diabetes that mimics the action of IAPP. As reflected by its name, IAPP forms amyloid fibrils in the islets of individuals with type 2 diabetes; whether this is a cause or a consequence of the islet dysfunction of type 2 diabetes is not yet clear.

Insulin secretion is a tightly regulated process designed to provide stable concentrations of glucose in blood during both fasting and feeding. This regulation is achieved by the coordinated interplay of various nutrients, GI hormones, pancreatic hormones, and autonomic neurotransmitters. Glucose, amino acids (arginine, etc.), fatty acids, and ketone bodies promote the secretion of insulin. Glucose is the primary insulin secretagogue, and insulin secretion is tightly coupled to the extracellular glucose concentration. Insulin secretion is much greater when the same amount of glucose is delivered orally compared to intravenously (incretin effect). Islets are richly innervated by both adrenergic and cholinergic nerves. Stimulation of α₂ adrenergic receptors inhibits insulin secretion, whereas β₂ adrenergic receptor agonists and vagal nerve stimulation enhance release. In general, any condition that activates the sympathetic branch of the autonomic nervous system (such as hypoxia, hypoglycemia, exercise, hypothermia, surgery, or severe burns) suppresses the secretion of insulin by stimulation of α₂ adrenergic receptors. Predictably, α₂ adrenergic receptor antagonists increase basal concentrations of insulin in plasma, and β₂ adrenergic receptor antagonists decrease them. Glucagon and somatostatin inhibit insulin secretion.

The pancreatic β cell is a highly specialized cell that has considerable structural and functional similarities to a sensory neuron: Both cell types quickly sense and respond to external stimuli. The molecular events controlling glucose-stimulated insulin secretion begin with the transport of glucose into the β cell via a facilitative glucose transporter (Figure 43–3). In rodents, this is the GLUT2, which has a distinctive, low affinity for glucose and is also the primary glucose transporter in hepatocytes. Human β cells express primarily GLUT1 but little GLUT2. Upon entry into the β cell, glucose is quickly phosphorylated by glucokinase (GK; hexokinase IV); this phosphorylation is the rate-limiting step in glucose metabolism in the β cell. GK's distinctive affinity for glucose leads to a marked increase in glucose metabolism over the range of 5-10 mM glucose, where glucose-stimulated insulin secretion is most pronounced. The glucose-6-phosphate produced by GK activity enters the glycolytic pathway, producing changes in NADPH and the ratio of ADP/ATP. Elevated ATP inhibits an ATP-sensitive K⁺ channel (K_ATP channel), leading to cell membrane depolarization. This K_ATP channel is a heteromeric protein that consists of an inward rectifying K⁺ channel (Kir6.2) and a closely associated protein known as the sulfonylurea receptor (SUR), which was identified originally because of its interaction with this class of drugs. Mutations in the K_ATP channel are responsible for some types of neonatal diabetes or hypoglycemia. Membrane depolarization then leads to opening of a voltage-dependent Ca²⁺ channel and increased intracellular Ca²⁺, resulting in exocytotic release of insulin from storage vesicles. These intracellular events are modulated by a number of processes, such as changes in cAMP production, amino acid metabolism, and the level of transcription factors. GPCRs for glucagon, GIP, and GLP-1 couple to G_s to stimulate adenylyl cyclase and insulin secretion; receptors for somatostatin and α₂ adrenergic agonists couple to G_i to reduce cellular cAMP production and secretion.

The pancreatic α cell, which has considerable similarity to the β cell, secretes glucagon, primarily in response to hypoglycemia. Glucagon biosynthesis begins with preproglucagon, which is processed in a cell-specific fashion to several biologically active peptides such as glucagon, GLP-1, and glucagon-like peptide-2 (GLP-2) (Figure 43-9). The molecular mechanisms that control glucagon secretion are unclear but may involve paracrine regulation by the β cell and neural input. In general, glucagon and insulin secretion are regulated in a reciprocal fashion; that is, the agents or processes that stimulate insulin secretion inhibit glucagon secretion. Notable exceptions are arginine and somatostatin: arginine stimulates and somatostatin inhibits the secretion of both hormones.

Figure 43–2.

Synthesis and processing of insulin. The initial peptide, preproinsulin (110 amino acids) consists of a signal peptide (SP), B chain, C peptide, and A chain. The SP is cleaved and S-S bonds form as the proinsulin folds. Two prohormone convertases, PC1 and PC2, cleave proinsulin into insulin, C peptide, and two dipeptides. Insulin and C peptide are stored in granules and co-secreted in equimolar quantities.

Figure 43–3.

Regulation of insulin secretion from a pancreatic β cell. The pancreatic β cell in a resting state (fasting blood glucose) is hyperpolarized. Glucose, entering via GLUT transporters (primarily GLUT1 in humans, GLUT2 in rodents), is metabolized and elevates cellular ATP, which inhibits. K⁺ entry through the K_ATP channel; the decreased K⁺ conductance results in depolarization, leading to Ca²⁺- dependent exocytosis of stored insulin. The K_ATP channel, actually a hetero-octamer composed of SUR1 and Kir 6.2 subunits, is the site of action of several classes of drugs: ATP binds to and inhibits Kir 6.2; sulfonylureas and meglitinides bind to and inhibit SUR1; all 3 agents thereby promote insulin secretion. Diazoxide and ADP-Mg²⁺ (low ATP) bind to and activate SUR1, thereby inhibiting insulin secretion. Incretins enhance insulin secretion.

Insulin Action. The insulin receptor is expressed on virtually all mammalian cell types, explaining the broad array of biological responses to insulin. The tissues that are considered critical for regulation of blood glucose are liver, skeletal muscle, and fat (Figure 43–1). However, recent evidence suggests that specific regions of the brain and the pancreatic islet are also important targets for insulin. Systemically, the actions of insulin are anabolic, and insulin signaling is critical for promoting the uptake, use, and storage of the major nutrients: glucose, lipids, and amino acids. Importantly, while insulin action stimulates glycogenesis, lipogenesis, and protein synthesis, it also inhibits the catabolism of these compounds. On a cellular level, insulin stimulates transport of substrates and ions into cells, promotes translocation of proteins between cellular compartments, regulates the action of specific enzymes, and controls gene transcription and mRNA translation. Some effects of insulin occur within seconds or minutes, such as activation of glucose and ion transport systems and phosphorylation or dephosphorylation of specific enzymes. Other effects, such as those promoting protein synthesis and regulating gene transcription, manifest over minutes to hours. The effects of insulin on cell proliferation and differentiation occur over days. Metabolic effects such as inhibition of lipolysis or hepatic glucose production occur rapidly, within minutes of increasing concentrations of plasma insulin; detectable increases in glucose clearance from the blood may take nearly an hour. The variability in the kinetics of insulin action probably relate to variable access to insulin receptors in different tissues, distinct intracellular signaling pathways, and the inherent kinetics of the various processes controlled by insulin.

The Insulin Receptor. Insulin action is transmitted through a receptor tyrosine kinase that bears functional similarity to the insulin-like growth factor 1 (IGF-1) receptor (Taniguchi et al., 2006). The insulin receptor is composed of linked (α/β subunit dimers that are products of a single gene; dimers linked by disulfide bonds form a transmembrane heterotetramer glycoprotein composed of two extracellular α-subunits and two membrane-spanning β-subunits (Figure 43–4). The number of receptors varies from as few as 40 per cell on the relatively insulin-insensitive erythrocytes, to 300,000 per cell on adipocytes and hepatocytes, cell types that are highly responsive to insulin.

Figure 43–4.

Pathways of insulin signaling. The binding of insulin to its plasma membrane receptor activates a cascade of downstream signaling events. Insulin binding activates the intrinsic tyrosine kinase activity of the receptor dimer, resulting in the tyrosine phosphorylation (Y-P) of the receptor's β subunits and a small number of specific substrates (yellow shapes): the Insulin Receptor Substrate (IRS) proteins, Gab-1 and SHC; within the membrane, a caveolar pool of insulin receptor phosphorylates caveolin (Cav), APS, and Cbl. These tyrosine-phosphorylated proteins interact with signaling cascades via SH2 and SH3 domains to mediate the effects of insulin, with specific effects resulting from each pathway. In target tissues such as skeletal muscle and adipocytes, a key event is the translocation of the Glut4 glucose transporter from intracellular vesicles to the plasma membrane; this translocation is stimulated by both the caveolar and non-caveolar pathways. In the non-caveolar pathway, the activation of PI3K is crucial, and PKB/Akt (anchored at the membrane by PIP3) and/or an atypical form of PKC is involved. In the caveolar pathway, caveolar protein flotillin localizes the signaling complex to the caveola; the signaling pathway involves series of SH2 domain interactions that add the adaptor protein CrkII, the guanine nucleotide exchange protein C3G, and small GTP-binding protein, TC10. The pathways are inactivated by specific phosphoprotein phosphatases (eg, PTB1B). In addition to the actions shown, insulin also stimulates the plasma membrane Na⁺,K⁺-ATPase by a mechanism that is still being elucidated; the result is an increase in pump activity and a net accumulation of K⁺ in the cell. Abbreviations: APS, adaptor protein with PH and SH2 domains; CAP, Cbl associated protein; CrkII, chicken tumor virus regulator of kinase II; GLUT4, glucose transporter 4; Gab-1, Grb-2 associated binder; MAP kinase, mitogen-activated protein kinase; PDK, phosphoinositide-dependent kinase; PI3 kinase, phosphatidylinositol-3-kinase; PIP3, phosphatidylinositol trisphosphate; PKB, protein kinase B (also called Akt); aPKC, atypical isoform of protein kinase C; Y, tyrosine residue; Y-P, phosphorylated tyrosine residue.

The α-subunits inhibit the inherent tyrosine kinase activity of the β-subunits. Insulin binding to the α-subunits releases this inhibition and allows transphosphorylation of one β-subunit by the other, and autophosphorylation at specific sites from the juxtamembrane region to the intracellular tail of the receptor. There are two splice variants of the α-subunits, which lead to insulin receptors with differential expression and ligand binding. In addition, insulin receptor dimers can form complexes with IGF-1 receptor α/β dimers, creating another distinct receptor isoform. Activation of the insulin receptor initiates signaling by phosphorylating a set of intracellular proteins such as the insulin receptor substrates (IRS) and Src-homology-2-containing protein (Shc). These insulin receptor substrates interact with effectors that amplify and extend the signaling cascade. As an example, binding of Shc to IRS1 leads to activation of the GTPase Ras and initiation of the protein kinase cascade involving MAPK and ERK that are involved in insulin-mediated gene transcription and cell growth.

Insulin action, at least for glucose transport, is critically dependent on the activation of phosphatidylinositol-3-kinase (PI3K). PI3K is activated by interaction with IRS proteins and generates phosphatidylinositol 3,4,5-trisphosphate (PIP3), which regulates the localization and activity of several downstream kinases, including Akt, atypical isoforms of protein kinase C (PKC ζ and λ/τ), and mammalian target of rapamycin (mTOR) (Huang and Czech, 2007). The isoform Akt2 appears to control the downstream steps that are important for glucose uptake in skeletal muscle and adipose tissue, and to regulate glucose production in the liver. Substrates of Akt2 coordinate the translocation of the glucose transporter 4 (GLUT4) to the plasma membrane through processes involving actin remodeling and other membrane trafficking systems (Zaid et al., 2008). Despite the centrality of PI3K/Akt2 in mediating insulin signaling in key target tissues, it seems likely that there are additional effects mediated by the insulin receptor that do not directly involve these enzymes. Actions of small G-proteins, such as Rac and TC10, have been implicated in the actin remodeling necessary for GLUT4 translocation.

GLUT4 is expressed in insulin-responsive tissues such as skeletal muscle and adipose tissue that constitute important sites of glucose disposal after meal ingestion. GLUT4 is one of a family of 13 glucose transporters in humans that share 12 membrane-spanning domains. GLUT4 is noteworthy among these transporters as the most dependent on discrete stimuli by insulin or other effectors; in the basal state, most GLUT4 resides in the intracellular space; following activation of insulin receptors, GLUT4 is shifted rapidly and in abundance to the plasma membrane, where it facilitates inward transport of glucose from the circulation. Insulin signaling also reduces GLUT4 endocytosis, increasing the residence time of the protein in the plasma membrane.

Following the facilitated diffusion into cells along a concentration gradient, glucose is phosphorylated to glucose-6-phosphate (G-6-P) by a family of hexokinases. Hexokinase II is found in association with GLUT4 in skeletal and cardiac muscle and in adipose tissue. Like GLUT4, hexokinase II is regulated transcriptionally by insulin. G-6-P is a branch-point substrate that can enter several pathways. G-6-P can be isomerized to G-1-P by phosphoglucomutase, and then the G-1-P can be stored as glycogen (insulin enhances the activity of glycogen synthase); G-6-P can enter the glycolytic pathway (leading to ATP production); G-6-P can also enter the pentose phosphate pathway.

PATHOPHYSIOLOGY AND DIAGNOSIS OF DIABETES MELLITUS

Glucose Homeostasis and the Diagnosis of Diabetes

Broad categories of glucose homeostasis as defined by the fasting blood glucose or the glucose level following an oral glucose challenge include:

Normal glucose homeostasis: Fasting plasma glucose <5.6 mmol/L (100 mg/dL)
Impaired fasting glucose (IFG): 5.6-6.9 mmol/L (100-125 mg/dL)
Impaired glucose tolerance (IGT): Glucose level between 7.8 and 11.1 mmol/L (140-199 mg/dL) 120 minutes after ingestion of 75 g liquid glucose solution
Diabetes mellitus (see Table 43–1)

Table 43-1Criteria for the Diagnosis of Diabetes

Table 43-1 Criteria for the Diagnosis of Diabetes

Symptoms of diabetes plus random blood glucose concentration ≥11.1 mM (200 mg/dL)^a or
Fasting plasma glucose ≥7.0 mM (126 mg/dL)^b or
Two-hour plasma glucose ≥11.1 mM (200 mg/dL) during an oral glucose tolerance test^c
HbA_1c ≥6.5%

^aRandom is defined as without regard to time since the last meal.

^bFasting is defined as no caloric intake for at least 8 h.

^cThe test should be performed using a glucose load containing the equivalent of 75 g anhydrous glucose dissolved in water; this test is not recommended for routine clinical use.

Note: In the absence of unequivocal hyperglycemia and acute metabolic decompensation, these criteria should be confirmed by repeat testing on a different day.

Adapted from Diabetes Care, 2010; 33:S62-S69.

The diagnosis of diabetes mellitus is currently based on the correlation of a diabetes-specific complication with a particular level of glycemia, that is, the level of glycemia at which a diabetes-specific complication like retinopathy begins to appear. The boundaries between normal glucose homeostasis, abnormal glucose homeostasis, and diabetes are described as abrupt transitions, but the data are best viewed as a spectrum.

Organizations such as the American Diabetes Association (ADA) and the World Health Organization (WHO) have adopted criteria for the diagnosis of diabetes, based on the fasting blood glucose, the glucose value following an oral glucose challenge, or the level of hemoglobin A_1c (HbA_1c; exposure of proteins to elevated [glucose] produces nonenzymatic glycation of proteins including Hb. Thus the level of HbA_1c represents a measure of the average glucose concentration to which the Hb has been exposed) (Table 43–1). Fasting plasma glucose is most widely used because of its convenience and low cost. The diagnostic criteria have recently (2009) been changed to also include a hemoglobin A_1c (A1C) value ≥6.5%. Impaired fasting glucose (IFG) and impaired glucose tolerance (IGT), previously termed prediabetes, or an A1C of 5.7-6.4% portend a markedly increased risk of progressing to type 2 diabetes. An A1C of 5.7-6.4%, IFG, and IGT do not identify the same group of individuals, but all are associated with increased risk of cardiovascular disease.

The four broad categories of diabetes include type 1 diabetes, type 2 diabetes, other forms of diabetes, and gestational diabetes (Table 43–2). Although hyperglycemia is common to all forms of diabetes, the pathogenic mechanisms leading to diabetes are quite distinct. As our current understanding of the molecular pathogenesis of the hyperglycemia improves, these categories will likely be subdivided into a number of other types of diabetes.

Table 43-2Different Forms of Diabetes Mellitus

Table 43-2 Different Forms of Diabetes Mellitus

Type 1 diabetes β-cell destruction, usually leading to absolute insulin deficiency)
1. Immune-mediated
2. Idiopathic
Type 2 diabetes (may range from predominantly insulin resistance with relative insulin deficiency to a predominantly insulin secretory defect with insulin resistance)
Other specific types of diabetes
1. Genetic defects of β cell function characterized by mutations in:
  1. Hepatocyte nuclear transcription factor (HNF) 4α (MODY 1)
  2. Glucokinase (MODY 2)
  3. HNF-1α (MODY 3)
  4. Insulin promoter factor-1 (IPF-1; MODY 4)
  5. HNF-1β (MODY 5)
  6. NeuroD1 (MODY 6)
  7. Mitochondrial DNA
  8. Subunits of ATP-sensitive K⁺ channel
  9. Proinsulin or insulin sequence conversion
2. Genetic defects in insulin action
  1. Type A insulin resistance
  2. Leprechaunism
  3. Rabson-Mendenhall syndrome
  4. Lipodystrophy syndromes
3. Diseases of the exocrine pancreas—pancreatitis, pancreatectomy, neoplasia, cystic fibrosis, hemochromatosis, fibrocalculous pancreatopathy, mutations in carboxyl ester lipase
4. Endocrinopathies—acromegaly, Cushing's syndrome, glucagonoma, pheochromocytoma, hyperthyroidism, somatostatinoma, aldosteronoma
5. Drug- or chemical-induced—Vacor (a rodenticide); see Table 43–3
6. Infections—congenital rubella, cytomegalovirus
7. Uncommon forms of immune-mediated diabetes—"stiff-person" syndrome, anti-insulin receptor antibodies
8. Other genetic syndromes sometimes associated with diabetes—Wolfram's syndrome, Down's syndrome, Klinefelter's syndrome, Turner's syndrome, Friedreich's ataxia, Huntington's chorea, Laurence-Moon-Biedlsyndrome, myotonic dystrophy, porphyria, Prader-Willi syndrome
Gestational diabetes mellitus (GDM)

MODY- maturity onset of diabetes of the young. Copyright 2010 American Diabetes Association. From Diabetes Care, Vol. 33 (Suppl 1), 2010; S62. Reprinted with permission from the American Diabetes Association.

Screening for Diabetes and Categories of Increased Risk Of Diabetes. Many individuals with type 2 diabetes are asymptomatic at the time of diagnosis, and diabetes is often found on routine blood testing as an outpatient or upon admission to the hospital for nonglucose-related reasons. A long, asymptomatic phase in type 2 diabetes (up to a decade) is likely responsible for the observation that up to 50% of individuals with diabetes may have a diabetes-related complication at the time of diagnosis. For these reasons, the ADA recommends widespread screening for type 2 diabetes of these individuals with the following features:

>45 years of age or
Body mass index >25 kg/m² with one of these additional risk factors: hypertension, low high-density lipoprotein, family history of type 2 diabetes, high-risk ethnic group (African American, Latino, Native American, Asian American, and Pacific Islander), abnormal glucose testing (IFG, IGT, A1C of 5.7-6.4%), cardiovascular disease, and women with polycystic ovary syndrome or who have previously delivered a large infant.

Earlier diagnosis and treatment of type 2 diabetes should delay diabetes-related complications and reduce the burden of the disease. Many also feel that earlier treatment may alter the natural history of the disease, but conclusive proof of this hypothesis is lacking. Likewise, informing individuals who are at increased risk for developing diabetes provides impetus to take steps to slow the progression to diabetes. A number of interventions including pharmacological agents and lifestyle modification are effective. Screening for type 1 diabetes is not currently recommended.

Pathogenesis of Type 1 Diabetes. Type 1 diabetes accounts for 5-10% of diabetes and results from autoimmune-mediated destruction of the β cells of the islet leading to total or near total insulin deficiency (von Herrath et al., 2007). Prior terminology included juvenile-onset diabetes mellitus or insulin-dependent diabetes mellitus. Although traditionally considered a disease of children and adolescents, type 1 diabetes resulting from autoimmune β cell destruction can occur at any age. In fact, in some reports, more than one-third of individuals developed type 1 diabetes after adolescence. Individuals with type 1 diabetes and their families have an increased prevalence of autoimmune diseases such as autoimmune adrenal insufficiency (Addison's disease), autoimmune thyroid disease (Graves' and Hashimoto's disease), pernicious anemia, vitiligo, and celiac sprue.

The concordance of type 1 diabetes in genetically identical twins is 40-60%, indicating a significant genetic component. The major genetic risk (40-50%) is conferred by HLA class II genes encoding HLA-DR and HLA-DQ (and possibly other genes with the HLA locus). Candidate gene association studies and recently genome-wide association studies have identified >20 additional loci that confer genetic susceptibility to type 1 diabetes (INS, PTPN22, CTLA4, and IL2RA, among others) (Concannon et al., 2009). However, there clearly is a critical interaction of genetics and an environmental or infectious agent. Most individuals with type 1 diabetes (~75%) do not have a family member with type 1 diabetes, and the genes conferring genetic susceptibility are found in a significant fraction of the nondiabetic population.

Genetically susceptible individuals are thought to have a normal β cell number or mass until β cell–directed autoimmunity develops and β cell loss begins. The first physiological abnormality that is detectable in affected subjects is loss of the first phase of glucose-stimulated insulin secretion. Prior to this, islet cell autoantibodies can be detected in the serum (known autoantigens include insulin, glutamate decarboxylase, protein tyrosine phosphatase IA-2[ICA-512], and zinc transporter 8 [SLC30A8]). The initiating or triggering stimulus for the autoimmune process is not known, but most favor exposure to viruses (enterovirus, etc.) or other ubiquitous environmental agents. Histological examinations of human pancreata during the prediabetic period and at presentation are quite limited, but animal models of type 1 diabetes show a T cell infiltrate with a predominance of CD8+ cells (insulitis). The β cell destruction is cell mediated, and there is also evidence that infiltrating cells produce local inflammatory agents such as TNF-α, IFN-γ, and IL-1, all of which can lead to β cell death. The β cell destruction occurs over a period of months to years and when >80% of the β cells are destroyed, hyperglycemia ensues and the clinical diagnosis of type 1 diabetes is made. Sometimes the clinical presentation of type 1 diabetes is with a concurrent illness (viral syndrome) and this may result in diabetic ketoacidosis; at other times, the typical symptoms of diabetes lead to the diagnosis. Most patients report several weeks of polyuria and polydipsia, fatigue, and often abrupt and significant weight loss. Some adults with the phenotypic appearance of type 2 diabetes (obese, not insulin-requiring initially) have islet cell autoantibodies suggesting autoimmune-mediated β cell destruction and are diagnosed as expressing latent-autoimmune diabetes of adults (LADA).

Pathogenesis of Type 2 Diabetes. The pathogenesis of type 2 diabetes mellitus is complex, and the condition is best thought of as a heterogeneous syndrome of dysregulated glucose homeostasis associated with impaired insulin secretion and insulin action (Das and Elbein, 2006; Doria et al., 2008; Taylor, 2008). Overweight or obesity is a common correlate of type 2 diabetes that occurs in ~ 80% of affected individuals. Increased lipid accumulation in depots in the abdomen, skeletal muscle cells, and hepatocytes has been linked to some of the common impairments. For the vast majority of persons developing type 2 diabetes, there is no clear inciting incident, but rather the condition is thought to develop gradually over years with progression through identifiable prediabetic stages.

In fundamental terms, type 2 diabetes results when there is insufficient insulin action to maintain plasma glucose levels in the normal range. Insulin action is the composite effect of plasma insulin concentrations (determined by islet β cell function) and insulin sensitivity of key target tissues (liver, skeletal muscle, and adipose tissue). These sites of regulation are all impaired to variable extents in patients with type 2 diabetes (Figure 43–5).

Figure 43–5.

Pathophysiology of type 2 diabetes mellitus. Graphs show data from diabetic (-) and non-diabetic (-) patients, comparing postprandial insulin and glucagon secretion and hepatic glucose production, and the sensitivities of muscle glucose use and adipocyte lipolysis to insulin.

The etiology of type 2 diabetes has a strong genetic component (Das and Elbein, 2006; Grant et al., 2009). It is a heritable condition with a relative 4-fold increased risk of disease for persons having a diabetic parent or sibling, increasing to 6-fold if both parents have type 2 diabetes. Consistent with this, the concordance rates of diabetes in monozygotic twins are 2- to 3-fold those in dizygotic twins. Based on linkage analysis, candidate gene searches, and genome-wide association studies, type 2 diabetes appears to be a complex multigenic condition, with many loci of susceptibility contributing to the ultimate phenotype. Although more than 20 genetic loci with clear associations to type 2 diabetes have been identified through recent genome-wide association studies, the contribution of each is relatively small. The genetic locus with the largest relative risk is the transcription factor TCF7L2. Of note, most of the genes currently associated with type 2 diabetes have a relevant connection to β cell function, with fewer linked to the action of insulin in target cells.

Impaired β Cell Function. In persons with type 2 diabetes, the sensitivity of the β cell to glucose is impaired, and there is also a loss of responsiveness to other stimuli such as insulinotropic GI hormones and neural signaling. This results in delayed secretion of insufficient amounts of insulin, allowing the blood glucose to rise dramatically after meals, and failure to restrain liver glucose release during fasting. Beyond the defect in functional properties of the β cell, the absolute mass of β cells is reduced in type 2 diabetes patients. It is estimated that persons with early type 2 diabetes have ~50% of the normal complement of β cells (Butler et al., 2003). This deficit is compounded by a gradual loss of β cell mass over time, potentially related to toxic effects of hyperglycemia. Progressive reduction of β cell mass and function explains the natural history of type 2 diabetes in most patients who require steadily increasing therapy to maintain glucose control.

Type 2 diabetic patients frequently have elevated levels of fasting insulin. This is not a reflection of accentuated β cell function but a result of their higher fasting glucose levels and insulin resistance. Another factor contributing to apparently high insulin levels early in the course of the disease is the presence of increased amounts of proinsulin. Proinsulin, the precursor to insulin, is inefficiently processed in the diabetic islet. Whereas healthy subjects have only 2-4% of total circulating insulin as proinsulin, type 2 diabetic patients can have 10-20% of the measurable plasma insulin in this form. Proinsulin has a considerably attenuated effect for lowering blood glucose compared to insulin.

Insulin Resistance. Insulin sensitivity is a quantifiable parameter that is measured as the amount of glucose cleared from the blood in response to a dose of insulin. The failure of normal amounts of insulin to elicit the expected response is referred to as insulin resistance. This is a relative term because there is inherent variability of insulin sensitivity among cells, tissues, and individuals. Insulin sensitivity is affected by many factors including age, body weight, physical activity levels, illness, and medications. Furthermore, insulin sensitivity varies within individuals over time and across groups or populations of subjects, even among healthy adults. Thus, insulin resistance is a relative designation but has considerable pathological significance because persons with type 2 diabetes or glucose intolerance have reduced responses to insulin and can easily be distinguished from groups with normal glucose tolerance.

The major insulin-responsive tissues are skeletal muscle, adipose tissue, and liver. Insulin resistance in muscle and fat is generally marked by a decrease in transport of glucose from the circulation. Hepatic insulin resistance generally refers to a blunted ability of insulin to suppress glucose production. Insulin resistance in adipocytes causes increased rates of lipolysis and release of fatty acids into the circulation, which can contribute to insulin resistance in liver and muscle, hepatic steatosis, and dyslipidemia. More generally the role of obesity in the etiology of type 2 diabetes is related to the insulin resistance in skeletal muscle and liver that comes with increased amounts of lipid storage, particularly in specific fat depots. The sensitivity of humans to the effects of insulin administration is inversely related to the amount of fat stored in the abdominal cavity; more visceral adiposity leads to more insulin resistance (Kahn, 2003). Similarly, intrahepatocyte or intramuscular fat, both commonly associated with obesity, are strongly linked to insulin resistance. Intracellular lipid or its byproducts may have direct effects to impede insulin signaling (Savage et al., 2007). Enlarged collections of adipose tissue, visceral or otherwise, is often infiltrated with macrophages and can become a site of chronic inflammation. Adipocytokines, secreted from adipocytes and immune cells, including TNF-α, IL-6, resistin, and retinol-binding protein 4, can also cause systemic insulin resistance. Insulin resistance is even more severe in obese persons with type 2 diabetes who have further reductions of insulin-stimulated glucose uptake into skeletal muscle and relatively low rates of nonoxidative glucose metabolism (glycogen synthesis), and impaired suppression of hepatic glucose production and adipocyte lipolysis, despite insulin concentrations that are often elevated.

Another important variable in determining insulin sensitivity is activity level. Sedentary persons are more insulin resistant than active ones, and physical training can improve insulin sensitivity. Physical activity can decrease the risk of developing diabetes and improve glycemic control in persons who have diabetes (Crandall et al., 2008). Insulin resistance is more common in the elderly; within populations, insulin sensitivity decreases linearly with age. Older individuals tend to be less physically active, which can contribute to insulin resistance. In addition, over the normal course of aging there is a decrease in muscle mass and an increase in fat mass, with an increased percentage of fat stored in the abdominal cavity.

At the cellular level, insulin resistance involves blunted steps in the cascade from the insulin receptor tyrosine kinase to translocation of GLUT4 transporters, but the molecular mechanisms are incompletely defined. There have been >75 different mutations in the insulin receptor discovered, most of which cause significant impairment of insulin action. These mutations affect insulin receptor number, movement to and from the plasma membrane, binding, and phosphorylation. Mutations involving the insulin binding domains of the extracellular α-chain cause the most severe syndromes, but specific variants in the intracellular portions of the receptor also cause severe insulin resistance. However, most insulin resistance associated with obesity and type 2 diabetes is not due to abnormalities of the insulin receptor. A central feature of the more common forms of insulin resistance is increased phosphorylation of serine, rather than tyrosine, in the insulin receptor and IRS proteins, inhibiting their activation and signaling. This process is mediated by protein serine/threonine kinases, which respond to an increased intracellular flux of fatty acids, specific lipid products, particularly diacylglycerol and ceramide, inflammatory mediators such as TNFα, and endoplasmic reticulum stress. In addition, chronic hyperinsulinemia, the typical correlate of insulin resistance, seems to increase IRS protein catabolism. It is known that insulin sensitivity is under genetic control, but it is unclear whether insulin-resistant individuals have mutations in specific components of the insulin signaling cascade or whether they have a complement of signaling effectors that operate at the lower range of normal. Regardless, it is apparent that insulin resistance clusters in families and is a major risk factor for the development of diabetes.

Dysregulated Hepatic Glucose Metabolism. In type 2 diabetes, hepatic glucose output is excessive in the fasting state and inadequately suppressed after meals. These are key components to the abnormal glycemic profile in diabetic patients, who have elevated glucose levels in the postabsorptive state and accentuated postprandial rises. Abnormal secretion of the islet hormones, both insufficient insulin and excessive glucagon, accounts for a significant portion of dysregulated hepatic glucose metabolism in type 2 diabetes. Increased concentrations of glucagon, especially in conjunction with hepatic insulin resistance, can lead to excessive hepatic gluconeogenesis and glycogenolysis and abnormally high fasting glucose concentrations.

The liver is also resistant to insulin action in type 2 diabetes. This contributes to the reduced potency of insulin to suppress hepatic glucose production and promote hepatic glucose uptake and glycogen synthesis after meals. Despite ineffective insulin effects on hepatic glucose metabolism, the lipogenic effects of insulin in the liver are maintained and even accentuated by fasting hyperinsulinemia. This contributes to hepatic steatosis and further worsening of insulin resistance.

Pathogenesis of Other Forms of Diabetes. Mutations in key genes involved in glucose homeostasis cause monogenic diabetes, which is inherited in an autosomal dominant fashion. These fall in two broad categories: diabetes onset in the immediate neonatal period (<6 months of age) and diabetes in children or adults (Murphy et al., 2008; Vaxillaire and Froguel, 2008). Some forms of neonatal diabetes are caused by mutations in the sulfonylurea receptor or its accompanying inward rectifying K⁺ channel and mutations in the insulin gene. Monogenic diabetes beyond the first year of life may appear clinically similar to type 1 or type 2 diabetes. In other instances, young individuals (adolescence to young adulthood) may have monogenic forms of diabetes known as maturity onset diabetes of the young (MODY). Phenotypically, these individuals are not obese and are not insulin resistant, or they may initially have modest hyperglycemia. The most common causes are mutations in key islet-enriched transcription factors or glucokinase (see discussion about molecular mechanisms of insulin secretion; Table 43–2 and Figure 43–3). Most individual with MODY are treated similarly to those with type 2 diabetes. Importantly, sulfonylureas may be a particularly effective treatment of individuals with MODY-3 or neonatal diabetes caused by a mutation in the K_ATP channel complex.

Diabetes may also be the result of other pathological processes such as acromegaly and Cushing's disease (Table 43–2). A number of medications promote hyperglycemia or lead to diabetes by either impairing insulin secretion or insulin action (Table 43–3). Most notable are asparaginase, glucocorticoids, pentamidine, nicotinic acid, α-interferon, protease inhibitors, and in particular, atypical antipsychotics (Chapter 16).

Epidemiology of Diabetes. The number of individuals with diabetes is increasing worldwide. Although the annual rate of type 1 diabetes has increased modestly over the past decade, the increased frequency of type 2 diabetes has been dramatic. In the U.S., 7-10% of adults are affected, with up to three times that many at risk. There is clear ethnic variability in the prevalence of type 2 diabetes. Among adults living in the U.S., there is a spectrum of prevalence with Native Americans (15%), Hispanics (12%), and African Americans (12%) having greater rates than persons of European ancestry (8%). While the upward trends in the industrialized world have been similar in many countries, there are now alarming increases in rates of type 2 diabetes worldwide. Most notable are China, India, and other countries in the Far East where the disease was previously uncommon (Chan et al., 2009). The largest increase in the number of people with diabetes is predicted to take place in the regions or countries with developing economies. Parallel increases in obesity rates and changes in diet and activity appear to be the major factors in this marked increase in diabetes.

Diabetes-Related Complications. Diabetes can cause metabolic derangements or acute complications, such as the life-threatening metabolic disorders of diabetic ketoacidosis and hyperglycemic hyperosmolar state. These require hospitalization for insulin administration, rehydration with intravenous fluids, and careful monitoring of electrolytes and metabolic parameters. Chronic complications of diabetes have been described in all forms of the disease and are commonly divided into microvascular and macrovascular complications. Microvascular complications occur only in individuals with diabetes, whereas macrovascular complications occur more frequently in individuals with diabetes but are not diabetes specific. The common microvascular complications include retinopathy, nephropathy and neuropathy. Macrovascular complications refer to increased atherosclerosis-related events such as myocardial infarction and stroke. The pharmacological management of diabetes-related complications is discussed elsewhere (e.g., Chapters 25, 28, 31, 64).

In the U.S., diabetes is the leading cause of blindness in adults, the leading reason for renal failure requiring dialysis or renal transplantation, and the leading cause of nontraumatic lower extremity amputations. Fortunately, most of these diabetes-related complications can be prevented, delayed, or reduced by near normalization of the blood glucose on a consistent basis. How chronic hyperglycemia causes these complications is unclear. For microvascular complications, current hypothesis are that hyperglycemia leads to advanced glycosylation end products (AGEs), increased glucose metabolism via the sorbitol pathway, increased formation of diacylglycerol leading to PKC activation, and increased flux through the hexosamine pathway. Growth factors such as vascular endothelial growth factor α may be involved in diabetic retinopathy and TGF-β in diabetic nephropathy.

Table 43-3Some Drugs that may Promote Hyperglycemiaor Hypoglycemia

Table 43-3 Some Drugs that may Promote Hyperglycemiaor Hypoglycemia

HYPERGLYCEMIA

HYPOGLYCEMIA

Glucocorticoids

β Adrenergic antagonists

Antipsychotics (atypical, others)

Ethanol Salicylates

Protease inhibitors

Non-steroidal anti-

β Adrenergic agonists

inflammatory drugs

Diuretics (thiazide, loop)

Pentamidine

Hydantoins (phenytoin, others)

ACE inhibitors Lithium chloride

Opioids (fentanyl, morphine, others)

Theophylline Bromocriptine

Diazoxide

Mebendazole

Nicotinic Acid

Pentamidine

Epinephrine

Interferons

Amphotericin B

Asparaginase

Acamprosate

Basiliximab

Thyroid hormones

For additional details, see Murad et al., 2009.

Therapy of Diabetes

Goals of Therapy. The goals of therapy for diabetes are to alleviate the symptoms related to hyperglycemia (fatigue, polyuria, etc.) and to prevent or reduce the acute and chronic complications of diabetes. Accomplishment of these goals requires a multidisciplinary team (physicians, nurse educators, pharmacists) with expertise in pharmacology, nutrition, and patient education. Central to the treatment plan is the patient who must actively participate in the care of his or her diabetes.

Glycemic control is assessed using both short-term (blood glucose self-monitoring) and long-term metrics (A1C, fructosamine). Using capillary blood glucose measurements, the patient assesses capillary blood glucose on a regular basis (fasting, before meals, or postprandially) and reports these values to the diabetes management team. A1C reflects glycemic control over the prior 3 months; glycosylated albumin (fructosamine) is a measure of glycemic control over the preceding 2 weeks.

Approaches to diabetes care are sometimes termed intensive insulin therapy, intensive glycemic control, and tight control. This chapter use the term comprehensive diabetes care to describe optimal therapy, which involves more than glucose management and includes aggressive treatment of abnormalities in blood pressure and lipids and detection and management of diabetes-related complications (Figure 43–6). Table 43–4 shows the ADA-recommended treatment goals for comprehensive diabetes care, for glucose, blood pressure, and lipids (Brunzell et al., 2008). Improved glycemic control reduces the complications when started relatively early in the course of both type 1 and type 2 diabetes, but very intensive glucose lowering (with A_1c near 6.0) has not shown benefit in individuals with type 2 diabetes and atherosclerotic disease (Duckworth et al., 2009; Holman et al., 2008; Skyler et al., 2009). A summary of available pharmacologic agents for the treatment of diabetes is at the end of this chapter (Table 43–9).

Nonpharmacologic Aspects of Diabetes Therapy. The patient with diabetes should be educated about nutrition, exercise, and medications aimed at lowering the plasma glucose. The role of the certified diabetes educator, a healthcare professional (nurse, dietician, or pharmacist) with specialized patient education skills, is critical. In terms of diet, the ADA uses the term medical nutrition therapy to describe the diet that coordinates calorie intake and other aspects of diabetes therapy such as pharmacological agents and exercise. In type 1 diabetes, matching caloric intake and insulin dosing is very important. In type 2 diabetes, the diet is directed at weight loss and reducing blood pressure and atherosclerotic risk. Exercise provides multiple benefits for patients with diabetes, but dosing of the glucose-lowering therapy may require adjustment to avoid exercise-related hypoglycemia.

In addition to lifestyle modification, the other major nonpharmacological means to reduce the progression of abnormal glucose metabolism is bariatric surgery (Sjostrom et al., 2004). A number of procedures, including gastric banding, gastric bypass, and biliopancreatic diversion, improve glucose tolerance and prevent or reverse type 2 diabetes.

Table 43-4Goals of Therapy in Diabetes

Table 43-4 Goals of Therapy in Diabetes

INDEX

GOAL^a

Glycemic control^b

A1C

<7.0%^c

Preprandial capillary plasma glucose

3.9-7.2 mmol/L (70-130 mg/dL)

Peak Postprandial capillary plasma glucose

10.0 mmol/L (<180 mg/dL)^d

Blood pressure

<130/80

Lipids^e

Low-density lipoprotein

<2.6 mmol/L (<100 mg/dL)^f

High-density lipoprotein

>1.1 mmol/L (>40 mg/dL)^g

Triglycerides

<1.7 mmol/L (<150 mg/dL)

^aAs recommended by the ADA goals should be individualized for each patient. Goals may be different forcertain patient populations.

^bA1C is primary goal.

^cWhile the ADA recommends an A1C <7.0% in general, in the individual patient it recommends that an approprprate goal for the individual patient based on age, duration of diabetes, life expectancy, other medical conditions, cardiovascular disease).

^dOne to two hours after beginning of a meal.

^eIn decreasing order of priority

^fIn individuals with conronary artery disease, an LDL <1.8 mmol (70 mg/dL) is the goal.

^gFor women, some suggest a goal that is 0.25 mmol/L (10 mg/dL) higher.

Adapted from Diabetes Care 33:S11, 2010.

Figure 43–6.

Components of comprehensive diabetes care.

Insulin Therapy

Insulin is the mainstay for treatment of virtually all type 1 and many type 2 diabetes patients. Insulin may be administered intravenously, intramuscularly, or subcutaneously. Long-term treatment relies predominantly on subcutaneous injection. Subcutaneous administration of insulin differs from physiological secretion of insulin in two major ways:

The absorption kinetics do not reproduce the rapid rise and decline of endogenous insulin in response to glucose following intravenous or oral administration.
Injected insulin is delivered into the peripheral circulation instead of being released into the portal circulation. Thus the portal/peripheral insulin concentration is not physiological, and this may alter the influence of insulin on hepatic metabolic processes.

Nonetheless, insulin delivered into the peripheral circulation can lead to normal or near-normal glycemia.

History. The discovery of insulin is a dramatic story. The discovery is attributed to Frederick Banting, Charles Best, J.J.R. Macleod, and J.B. Collip at the University of Toronto, but others provided important observations and techniques that made it possible (Bliss, 2005). In 1869, a German medical student, Paul Langerhans, noted that the pancreas contains two distinct groups of cells—the acinar cells, which secrete digestive enzymes, and cells that are clustered in islands, or islets, which he suggested served a second function. Direct evidence for this function came in 1889, when Minkowski and von Mering showed that pancreatectomized dogs exhibit a syndrome similar to human diabetes mellitus. Thereafter, there were numerous attempts to extract the pancreatic substance responsible for regulating blood glucose. Between 1903 and 1909, the Romanian physiologist Nicolas Paulesco found that injections of pancreatic extracts reduced urinary sugar and ketones in diabetic dogs and published results clearly indicating isolation of a glucose-lowering compound. The significance of these findings was recognized only years later.

Unaware of much of this work, Frederick Banting, a young Canadian surgeon, convinced J.J.R. Macleod, a professor of physiology in Toronto, to allow him access to a laboratory to search for the antidiabetic principle of the pancreas. Banting began work on the assumption that the islets secreted a glucose-lowering hormone that was destroyed by proteolytic digestion prior to or during extraction. Together with Charles Best, a fourth-year medical student, he attempted to overcome the problem by ligating the pancreatic ducts so that the acinar tissue degenerated, leaving the islets undisturbed. Using an ethanol and acid extraction procedure, Banting and Best obtained a pancreatic isolate that decreased the concentration of blood glucose in diabetic dogs. Under the guidance of Macleod, Banting and Best received help from J.B. Collip, a chemist with expertise in extraction and purification of epinephrine. The first patient to receive the active extracts was Leonard Thompson, age 14. He presented at the Toronto General Hospital with a blood glucose level of 500 mg/dL (28 mM). Despite rigid control of his diet (450 kcal/day), he continued to excrete large quantities of glucose, and, without insulin, would likely have died within a few months. The administration of the pancreatic extracts reduced the plasma concentration and urinary excretion of glucose, and the patient demonstrated marked clinical improvement. Replacement therapy with the newly discovered hormone, insulin, had interrupted what was clearly an otherwise fatal metabolic disorder. Stable extracts eventually were obtained, and patients in many parts of North America soon were being treated with insulin from porcine and bovine sources. The first of several Nobel Prizes associated with insulin was awarded to Banting and Macleod in 1923. A furor over credit followed immediately, and Banting announced that he would share his prize with Best; Macleod did the same with Collip (Bliss, 2005). Frederick Sanger established the amino acid sequence of insulin and received the Nobel Prize in 1958. Dorothy Hodgkin (Nobel prize for chemistry, 1964) and coworkers elucidated insulin's three-dimensional structure. Insulin was the hormone for which Yalow and Berson first developed the radioimmunoassay, which was recognized with the Nobel Prize in 1977.

Insulin Preparation and Chemistry. Human insulin, produced by recombinant DNA technology, is soluble in aqueous solution. Most preparations are supplied at neutral pH, which improves stability and permits short-term storage at room temperature. Beginning shortly after its discovery in 1921, patients were treated with insulin preparations prepared from porcine or bovine pancreatic extracts for >70 years. With the advent of human insulin, beef and pork insulin are no longer produced. A number of other insulin preparations that once were widely used such as lente, ultralente, and protamine zinc insulin are no longer available.

Doses and concentration of clinically used insulin preparations are expressed in international units. This tradition dates to a time when preparations of hormones were impure and the concentration of the hormone was standardized by bioassay. One unit of insulin is defined as the amount required to reduce the blood glucose concentration in a fasting rabbit to 45 mg/dL (2.5 mM). Commercial preparations of insulin are supplied in solution or suspension at a concentration of 100 units/mL, which is ~3.6 mg insulin per milliliter (0.6 mM) and termed U-100. Insulin also is available in a more concentrated solution (500 units/mL or U-500) for patients who are resistant to the hormone. In the past, other concentrations of insulin such as U-40 were available. Other insulin formulations such as inhaled insulin and injected human proinsulin were tested but either were not clinically useful or were used for only a brief period of time. Work is ongoing to develop delivery approaches that do not involve injection.

Insulin Formulations. Preparations of insulin are classified according to their duration of action into short-acting and long-acting (Table 43–5). Within the short-acting acting category, some distinguish the very rapid-acting insulins (aspart, glulisine, lispro) from regular insulin. Likewise, some distinguish formulations with a longer duration of action (detemir, glargine) from NPH insulin. Two approaches are used to modify the absorption and pharmacokinetic profile of insulin. The first approach, which has been used for >70 years to alter the absorption profile of native insulin, is based on formulations that slow the absorption following subcutaneous injection. The other approach is to alter the amino acid sequence or protein structure of human insulin so that it retains its ability to bind the insulin receptor, but its behavior in solution or following injection is either accelerated or prolonged in comparison to native or regular insulin (Figure 43–7).

Table 43-5Properties of Insulin Preparations

Table 43-5 Properties of Insulin Preparations

TIME OF ACTION

PREPARATION

ONSET, h

PEAK, h

EFFECTIVE DURATION, h

Short-acting

Aspart

<0.25

0.5-1.5

3-4

Glulisine

<0.25

0.5-1.5

3-4

Lispro

<0.25

0.5-1.5

3-4

Regular

0.5-1.0

2-3

4-6

Long-acting

Detemir

1-4

—^a

20-24

Glargine

1-4

—^a

20-24

NPH

1-4

6-10

10-16

Insulin combinations

75/25-75% protamine lispro, 25% lispro

<0.25

1.5 h

Up to 10-16

70/30-70% protamine aspart, 30% aspart

<0.25

1.5 h

Up to 10-16

50/50-50% protamine lispro, 50% lispro

<0.25

1.5 h

Up to 10-16

70/30-70% NPH, 30% regular

0.5-1

Dual^b

10-16

^aGlargine and detemir have minimal peak activity.

^bDual: two peaks - one at 2-3 h; the second one several hours later.

Copyright 2004 American Diabetes Association. Adapted with permission from Skyler JS. Insulin treatment. In HE Lebovitz, ed., Therapy for Diabetes Mellitus. American Diabetes Association, Alexandria, VA, 2004.

Figure 43–7.

Insulin analogs. Modifications of native insulin can alter its pharmacokinetic profile. Reversing amino acids 28 and 29 in the B chain (lispro) or substituting Asp for Pro^28B (aspart) gives analogs with reduced tendencies for molecular self-association that are faster acting. Altering Asp^3B to Lys and Lys^29B to Glu produces an insulin (glulisine) with a more rapid onset and a shorter duration of action. Substituting Gly for Asn^21A and lengthening the B chain by adding Arg³¹ and Arg³² produces a derivative (glargine) with reduced solubility at pH 7.4 that is, consequently, absorbed more slowly and acts over a longer period of time. Deleting Thr^30B and adding a myristoyl group to the ∊-amino group of Lys^29B (detemir) enhances reversible binding to albumin, thereby slowing transport across vascular endothelium to tissues and providing prolonged action.

The wide variability in the kinetics of insulin action between and even within individuals must be emphasized. The time to peak hypoglycemic effect and insulin levels can vary by 50%. This variability is caused, at least in part, by large variations in the rate of subcutaneous absorption.

Short-Acting Regular Insulin. Native or regular insulin molecules associate as hexamers in aqueous solution at a neutral pH and this aggregation slows absorption following subcutaneous injection. Regular insulin should be injected 30-45 minutes before a meal. Regular insulin also may be given intravenously or intramuscularly.

Short-Acting Insulin Analogs. The development of short-acting insulin analogs that retain a monomeric or dimeric configuration is a major advance in insulin therapy (Figures 43–7 and 43–8; Table 43–5). These analogs are absorbed more rapidly from subcutaneous sites than regular insulin. Consequently, there is a more rapid increase in plasma insulin concentration and an earlier response. Insulin analogs should be injected ≤15 minutes before a meal.

Insulin lispro (humalog) is identical to human insulin except at positions B28 and B29, where the sequence of the two residues has been reversed to match the sequence in IGF-1 (which does not self-associate). Like regular insulin, lispro exists as a hexamer in commercially available formulations. Unlike regular insulin, lispro dissociates into monomers almost instantaneously following injection. This property results in the characteristic rapid absorption and shorter duration of action compared with regular insulin. Two therapeutic advantages have emerged with lispro as compared with regular insulin. First, the prevalence of hypoglycemia is reduced with lispro; second, glucose control, as assessed by A1C, is modestly but significantly improved (0.3-0.5%).

Insulin aspart (novolog) is formed by the replacement of proline at B28 with aspartic acid. This reduces self-association to a degree similar to lispro. Like lispro, insulin aspart dissociates rapidly into monomers following injection. Comparison of a single subcutaneous dose of aspart and lispro in a group of type 1 diabetes patients revealed similar plasma insulin profiles. In clinical trials, insulin aspart and insulin lispro have had similar effects on glucose control and hypoglycemia frequency, with lower rates of nocturnal hypoglycemia as compared with regular insulin.

Insulin glulisine (apidra) is formed when glutamic acid replaces lysine at B29, and lysine replaces asparagine at B23; these substitutions result in a reduction in self-association and rapid dissociation into active monomers. The time–action profile of insulin glulisine is similar to that of insulin aspart and insulin lispro. Similar to insulin aspart, insulin glulisine has been approved in the U.S. for continuous subcutaneous insulin infusion (CSII) pump use.

Long-Acting Insulins. Neutral protamine hagedorn (NPH; insulin isophane) is a suspension of native insulin complexed with zinc and protamine in a phosphate buffer. This produces a cloudy or whitish solution in contrast to the clear appearance of other insulin solutions. Because of this formulation, the insulin dissolves more gradually when injected subcutaneously and thus its duration of action is prolonged. NPH insulin is usually given either once a day (at bedtime) or twice a day in combination with short-acting insulin. In patients with type 2 diabetes, long-acting insulin is often given at bedtime to help normalize fasting blood glucose. It should be noted that the use of long-acting basal insulin alone will not control postprandial glucose elevation in insulin-deficient type 1 or type 2 diabetes.

Insulin glargine (lantus) is a long-acting analog of human insulin that is produced following two alterations of human insulin. Two arginine residues are added to the C terminus of the B chain, and an asparagine molecule in position 21 on the A chain is replaced with glycine. Insulin glargine is a clear solution with a pH of 4.0, which stabilizes the insulin hexamer. When injected into the neutral pH of the subcutaneous space, aggregations occurs, resulting in prolonged, but predictable, absorption from the injection site. Owing to insulin glargine's acidic pH, it cannot be mixed with short-acting insulin preparations (i.e., regular insulin, aspart, or lispro) that are formulated at a neutral pH.

In clinical studies glargine has a sustained peakless absorption profile, and provides a better once-daily 24-hour insulin coverage than NPH insulin. Evidence from clinical trials also suggests that glargine has a lower risk of hypoglycemia, particularly overnight compared to NPH insulin. Glargine may be administered at any time during the day with equivalent efficacy and does not accumulate after several injections. Glargine has been shown in clinical studies to normalize fasting (postabsorptive) glucose levels following once-daily administration in patients with type 2 diabetes. Sometimes, splitting the dose of glargine may be needed in very insulin-sensitive type 1 diabetes patients to achieve fasting (basal) glucose levels in the target range and avoid hypoglycemia. Unlike traditional insulin preparations that are absorbed more rapidly from the abdomen than from the arm or leg, the site of administration does not influence the time–action profile of glargine. Similarly, exercise does not influence glargine's unique absorption kinetics, even when the insulin is injected into a working limb.

Glargine binds with a slightly greater affinity to IGF-1 receptors as compared with human insulin. However, this slightly increased binding is still approximately two log scales lower than that of IGF-1. Controversy currently exists whether glargine insulin is associated with an increased chance of malignancy or an acceleration of underlying malignancy, but most feel that the available evidence is circumstantial and not sufficiently convincing to change prescribing patterns (Pocock and Smeeth, 2009; Smith and Gale, 2009; Weinstein et al., 2009).

Insulin determir (levemir) is an insulin analog modified by the addition of a saturated fatty acid to the ∊ amino group of LysB29, yielding a myristoylated insulin. When insulin detemir is injected subcutaneously, it binds to albumin via its fatty acid chain. Clinical studies in patients with type 1 diabetes have demonstrated that when insulin detemir is administered twice a day, it has a smoother time–action profile and produces a reduced prevalence of hypoglycemia than NPH insulin. The absorption profiles of glargine and determir insulin are similar, but detemir often requires twice-daily administration.

Other Insulin Formulations. Stable, mixed combinations (Table 43–5) of NPH and regular insulin in proportions of 70:30 combinations are available. Combinations of lispro protamine/lispro (50/50 and 75/25) and aspart protamine/aspart (70/30) are also available in the U.S.

Figure 43–8.

Commonly used insulin regimens. Panel A shows administration of a long-acting insulin like glargine (detemir could also be used but often requires twice-daily administration) to provide basal insulin and a pre-meal short-acting insulin analog (Table 43–5). Panel B shows a less intensive insulin regimen with BID injection of NPH insulin providing basal insulin and regular insulin or an insulin analog providing meal-time insulin coverage. Only one type of shorting-acting insulin would be used. Panel C shows the insulin level attained following subcutaneous insulin (short-acting insulin analog) by an insulin pump programmed to deliver different basal rates. At each meal, an insulin bolus is delivered. B=breakfast; L=lunch; S=supper; HS=bedtime. Upward arrow shows insulin administration at mealtime. (Copyright 2008 American Diabetes Association. From Kaufman FR (ed), Medical Management of Type 1 Diabetes, Fifth Edition. Modified with permission from the American Diabetes Association.)

Insulin Delivery. Most insulin is injected subcutaneously. Pen devices containing prefilled regular, lispro, NPH, glargine, premixed lispro protamine-lispro, or premixed aspart protamine-aspart have proven to be popular with many diabetic patients. Jet injector systems that enable patients to receive subcutaneous insulin injections without a needle are available. Intravenous infusions of insulin are useful in patients with ketoacidosis or when requirements for insulin may change rapidly, such as during the perioperative period, during labor and delivery, and in intensive care situations.

Continuous Subcutaneous Insulin Infusion. Short-acting insulins are the only form of the hormone used in subcutaneous infusion pumps. A number of pumps are available for continuous subcutaneous insulin infusion (CSII) therapy. CSII, or pump, therapy is not suitable for all patients because it demands considerable attention. However, for patients interested in intensive insulin therapy, a pump may be an attractive alternative to several daily injections. Insulin pumps provide a constant basal infusion of insulin and have the option of different infusion rates during the day and night to help avoid the dawn phenomenon (rise in blood glucose that occurs just prior to awakening from sleep) and bolus injections that are programmed according to the size and nature of a meal.

Glucose sensors that measure the interstitial glucose are now available and may be helpful in patients with labile blood glucose and frequent hypoglycemia (Juvenile Diabetes Research Foundation Continuous Glucose Monitoring Study Group, 2008). Although patients can use both a pump and a sensor, these devices are not yet integrated into a closed-loop system. Similar to the collected evidence with CSII, the benefits of continuous glucose monitoring have not been conclusively demonstrated, nor have clear selection criteria for appropriate patient groups been developed. However, the results from recent studies suggest that real-time glucose monitoring may lead to modest improvements in glycemic control and improved patient/family quality of life.

Pump therapy presents some unique problems. Because all the insulin used is short acting and there is a minimal amount of insulin in the subcutaneous pool at any given time, insulin deficiency and ketoacidosis may develop rapidly if therapy is interrupted accidentally. Although modern pumps have warning devices that detect changes in line pressure, mechanical problems such as pump failure, dislodgement of the needle, aggregation of insulin in the infusion line, or accidental kinking of the infusion catheter may occur. There also is a possibility of subcutaneous abscesses and cellulitis. Selection of the most appropriate patients is extremely important for success with pump therapy. Offsetting these potential problems, pump therapy is capable of producing a more physiological profile of insulin replacement during exercise (where insulin production is decreased) and therefore less hypoglycemia than do traditional subcutaneous insulin injections.

Factors That Affect Insulin Absorption. Factors that determine the rate of absorption of insulin after subcutaneous administration include the site of injection, the type of insulin, subcutaneous blood flow, smoking, regional muscular activity at the site of the injection, the volume and concentration of the injected insulin, and depth of injection (insulin has a more rapid onset of action if delivered intramuscularly rather than subcutaneously). Increased subcutaneous blood flow (brought about by massage, hot baths, or exercise) increases the rate of absorption.

Insulin usually is injected into the subcutaneous tissues of the abdomen, buttock, anterior thigh, or dorsal arm. Absorption is usually most rapid from the abdominal wall, followed by the arm, buttock, and thigh. If a patient is willing to inject into the abdomen, injections can be rotated throughout the entire area, thereby eliminating the injection site as a cause of variability in the rate of absorption. The abdomen currently is the preferred site of injection in the morning because insulin is absorbed 20-30% faster from that site than from the arm. If the patient refuses to inject into the abdominal area, it is preferable to select a consistent injection site for each component of insulin treatment (e.g., before-breakfast dose into the thigh, evening dose into the arm). Rotation of insulin injection sites traditionally has been advocated to avoid lipohypertrophy or lipoatrophy, although these conditions are infrequent with current preparations of insulin. In a small group of patients, subcutaneous degradation of insulin has been observed, and this has necessitated the injection of large amounts of insulin for adequate metabolic control.

Insulin Dosing and Regimens. In most patients, insulin-replacement therapy includes long-acting insulin (basal) and a short-acting insulin to provide postprandial needs. In a mixed population of type 1 diabetes patients, the average dose of insulin is usually 0.6-0.7 units/kg body weight per day, with a range of 0.2-1 units/kg per day. Obese patients generally and pubertal adolescents require more (~1-2 units/kg per day) because of resistance of peripheral tissues to insulin. Patients who require less insulin than 0.5 units/kg per day may have some endogenous production of insulin or may be more sensitive to the hormone because of good physical conditioning. The basal dose suppresses lipolysis, proteolysis, and hepatic glucose production; it is usually 40-50% of the total daily dose with the remainder as prandial or pre-meal insulin. The insulin dose at meal time should reflect the anticipated carbohydrate intake (many patients with type 1 diabetes calculate a ratio of the insulin dose to the number of grams of carbohydrate). A supplemental scale of short-acting insulin is added to the prandial insulin dose to allow correction of the BG. With all insulin dosing, the provider should consider the insulin sensitivity of the patient and adjust the insulin dosing accordingly. Insulin administered as a single daily dose of long-acting insulin, alone or in combination with short-acting insulin, is rarely sufficient to achieve euglycemia. More complex regimens that include multiple injections of long-acting or short-acting insulin are needed to reach this goal. In all patients, careful monitoring of therapeutic end points directs the insulin dose used. This approach is facilitated by self-glucose monitoring and measurements of A1C.

A number of commonly used dosage regimens that include mixtures of insulin given in two or more daily injections are depicted in Figure 43–8. An effective regimen involving multiple daily injections consisting of basal administration of long-acting insulin (e.g., insulin glargine or determir) either before breakfast or at bedtime and preprandial injections of a short-acting insulin (Weng et al., 2008). This method is called basal/bolus and is very similar to the pattern of insulin administration achieved with a subcutaneous infusion pump. Another regimen used is the split-mixed regimen involving the pre-breakfast and pre-supper injection of a mixture of short- and long-acting insulins. When the pre-supper NPH insulin is not sufficient to control hyperglycemia throughout the night, the evening dose may be divided into a pre-supper dose of regular insulin followed by NPH insulin at bedtime.

Individuals with diabetes may sometimes consume smaller amounts of food than originally planned. This, in the presence of a previously injected dose of insulin that was based on anticipation of a larger meal, could result in postprandial hypoglycemia. Thus, in patients who have gastroparesis or loss of appetite, injection of a short-acting analog postprandially, based on the amount of food actually consumed, may provide smoother glycemic control.

Adverse Events. The most common adverse reaction during insulin therapy is hypoglycemia. Hypoglycemia is the major risk that must be weighed against benefits of efforts to normalize glucose control. Additional information about hypoglycemia and its treatment is provided later in this chapter. Insulin is an anabolic hormone, and insulin treatment of both type 1 and type 2 diabetes is associated with modest weight gain. Paradoxically, improved glycemic control may initially lead to the deterioration of retinopathy in rare patients, but this is followed by a long-term reduction in diabetes-related complications. There has been a dramatic decrease in the incidence of allergic reactions to insulin with the transition to recombinant human insulin; these may still occur as a result of reaction to the small amounts of aggregated or denatured insulin in preparations, to minor contaminants, or because of sensitivity to one of the components added to insulin in its formulation (protamine, Zn²⁺, etc.). Human insulin, as delivered to patients with diabetes, is immunogenic as reflected by the observation that many patients have circulating anti-insulin antibodies, but these do not alter insulin pharmacokinetics or action. Atrophy of subcutaneous fat at the site of insulin injection (lipoatrophy) was a rare side effect of older insulin preparations. Lipohypertrophy (enlargement of subcutaneous fat depots) has been ascribed to the lipogenic action of high local concentrations of insulin.

Insulin Treatment of Ketoacidosis and Other Special Situations. Acutely ill diabetic patients may have metabolic disturbances that are sufficiently severe or labile to justify intravenous administration of insulin (Kitabchi et al., 2009). Such treatment is most appropriate in patients with ketoacidosis or severe hyperglycemia with a hyperosmolar state. Insulin infusion inhibits lipolysis and gluconeogenesis completely and produces near-maximal stimulation of glucose uptake. In most patients with diabetic ketoacidosis, blood glucose concentrations will fall by ~10% per hour; the acidosis is corrected more slowly. As treatment proceeds, it often is necessary to administer glucose along with the insulin to prevent hypoglycemia but to allow clearance of all ketones. Some physicians prefer to initiate therapy with a loading dose of insulin, but this tactic appears unnecessary because steady-state concentrations of the hormone are achieved within 30 minutes with a constant infusion. Patients with nonketotic hyperglycemic hyperosmolar state may be more sensitive to insulin than are those with ketoacidosis. Appropriate replacement of fluid and electrolytes is an integral part of the therapy in both situations because there is always a major deficit. Regardless of the insulin regimen, the key to effective therapy is careful and frequent monitoring of the patient's clinical status, glucose, and electrolytes. A frequent error in the management of such patients is the failure to administer long-acting insulin subcutaneously before the insulin infusion is discontinued.

Treatment of Diabetes in Children or Adolescents. Diabetes is one of the most common chronic diseases of childhood, and rates of type 1 diabetes in American youth are estimated at 1 in 300, with an increasing incidence over the past 20 years. One of the unfortunate corollaries of the growing rates of obesity over the past three decades is an increase in the numbers of children and adolescents with nonautoimmune, or type 2, diabetes. Current estimates are that 15-20% of new cases of pediatric diabetes may in fact be type 2 diabetes; rates vary by ethnicity, with disproportionately high rates in Native Americans, African Americans, and Latinos. Because of a paucity of clinical trials performed in children, there is only limited information on which to base decisions for appropriate therapy. Thus the results in the Diabetes Control and Complications Trial with young and middle-aged adults have been extrapolated to the pediatric population such that current practice is for more intensive, physiologically based insulin replacement with a goal of tight glucose control (Diabetes Control Complications Trial Research Group, 1993). This is achieved with combinations of basal and prandial insulin replacement. The primary limiting factor of more aggressive insulin therapy is hypoglycemia, a problem with special concerns in young children. Diabetic patients <5 years old are at greater risk for hypoglycemia, have increased rates of severe hypoglycemia with seizures and coma, and may suffer permanent cognitive dysfunction as a result of repeated episodes of low blood glucose. Older children and adolescents do not seem to have demonstrable cognitive impairment related to hypoglycemia; good glycemic control is associated with better mental function (Kodl and Seaquist, 2008).

The treatment of type 1 diabetes in children and adolescents has changed with the availability of newer technologies. The standard for insulin treatment now includes multiple dose regimens with three to five injections per day or CSII. Split/mixed regimens using NPH and regular insulin have been increasingly supplanted by regimens using insulin analogs because they offer more flexibility in dosing and meal patterns. Similarly, CSII is used with increasing frequency in the pediatric diabetic population. Moreover, recent studies show that this approach is applicable to young children as well as older children and adolescents.

Because of the nearly uniform association of type 2 diabetes with obesity in the pediatric age group, lifestyle management is the recommended first step in therapy. Goals of reducing body weight while maintaining normal linear growth, and increasing physical activity, are broadly recommended and can be effective when patients are compliant. There have been few clinical trials of glucose lowering therapy in pediatric type 2 diabetes. The only medication currently approved by the FDA specifically for medical treatment of type 2 diabetes is metformin. Metformin is approved for children as young as 10 years of age and is available in a liquid formulation (100 mg/mL). Results from clinical trials have shown that both metformin and glimepiride effectively lower blood glucose in affected patients. Insulin is the typical second line of therapy after metformin; basal insulin can be added to oral agent therapy or multiple daily injections can be used when simpler regimens are not successful. Weight gain is a more significant problem than hypoglycemia with insulin treatment in pediatric type 2 diabetes. Other diabetes medications, such as thiazolidinediones, α-glucosidase inhibitors, DPP-4 inhibitors, and exenatide, have been tried empirically in type 2 diabetic adolescents, but there is no systematically collected data on efficacy or safety of these agents in the pediatric population.

Management of Diabetes in Hospitalized Patients. Hyperglycemia is common in hospitalized patients. Depending on how hyperglycemia is defined, prevalence estimates of elevated blood glucose among inpatients with and without a prior diagnosis of diabetes range between 20% and 100% for patients treated in intensive care units (ICUs) and 30% and 83% outside the ICU. Although most of these individuals will have known diabetes, ~30% of hospitalized patients will have elevated blood glucose levels without a prior diagnosis of diabetes (Falciglia, 2007). Patients admitted to the hospital often have a number of challenges to glucose regulation in addition to those faced by diabetic outpatients (Donner and Flammer, 2008). Stress of illness has been associated with insulin resistance, possibly the result of counterregulatory hormone secretion, cytokines, and other inflammatory mediators. Food intake is often variable due to concurrent illness or preparation for diagnostic testing. Medications used in the hospital, such as glucocorticoids or dextrose-containing intravenous solutions, can exacerbate tendencies toward hyperglycemia. Finally fluid balance and tissue perfusion can affect the absorbance of subcutaneous insulin and the clearance of glucose. Therapy of hyperglycemia in hospitalized patients needs to be adjusted for these variables.

There is an emerging body of information indicating that hyperglycemia portends poor outcomes in hospitalized patients, most notably in the critically ill, and that this effect is independent of severity of illness (Finfer et al., 2009; van den Berghe et al., 2001). The mechanisms for this association have not been fully explained, and controversy persists about the optimal level of glycemia in hospitalized patients. The ADA currently suggests these blood glucose targets: 140-180 m/dL (7.8-10.0 mM) in critically ill patients and random glucose of 180 mg/dL (10 mM) or pre-meal glucose of 140 mg/dL (7.8 mM) in noncritically ill patients. Great emphasis should be placed on steps taken to minimize hypoglycemia in both settings.

Insulin is the cornerstone of treatment of hyperglycemia in hospitalized patients (Moghissi et al., 2009). For critically ill patients and those with variable blood pressure, edema, and tissue perfusion, intravenous insulin is the treatment of choice. This method of insulin administration has been firmly established in the care of critically ill patients with elevated blood glucose and provides the most flexible and precise means of treatment. A number of algorithms have been adapted to allow rapid titration with adjustments to maintain blood glucose in a target range. For patients who are more stable, subcutaneous insulin regimens using combinations of basal and prandial insulin is the standard. There is considerable evidence that reactive treatment, using sliding scale regimens, is inferior and associated with wider fluctuations in blood glucose and greater rates of both hyper- and hypoglycemia. Oral agents have a limited place in treatment of hyperglycemic patients in the hospital because of slow onset of action, insufficient potency, need for intact GI function, and side effects. In general, oral glucose-lowering medications should be discontinued on hospital admission and can be restarted at discharge.

Intravenous administration of insulin also is well suited to the treatment of diabetic patients during the perioperative period and during childbirth. There is debate, however, about the optimal route of insulin administration during surgery. Although some clinicians advocate subcutaneous insulin administration, most recommend intravenous insulin infusion. Some physicians give patients half their normal daily dose of insulin as long-acting insulin subcutaneously on the morning of an operation and then administer 5% dextrose infusions during surgery to maintain glucose concentrations. This approach provides less minute-to-minute control than is possible with intravenous regimens and also may increase the likelihood of hypoglycemia. For patients with type 1 diabetes, failure to provide some basal insulin at all times can precipitate diabetic ketoacidosis.

INSULIN SECRETAGOGUES AND ORAL HYPOGLYCEMIC AGENTS

A variety of sufonylureas, meglitinides, GLP-1 agonists, and inhibitors of dipeptidyl peptidase-4 (DPP-4) are used as secretagogues to stimulate insulin release (Table 43–6).

Table 43-6Properties of Insulin Secretagogues

Table 43-6 Properties of Insulin Secretagogues

CLASS/GENERIC NAME

DAILY DOSAGE^a, mg

DURATION OF ACTION, h

Sulfonylureas - first generation

Chlorpropamide

100-500

>48

Tolazamide

100-1000

12-24

Tolbutamide

1000-3000

6-12

Sulfonylureas - second generation

Glimepiride

1-8

Glipizide

5-40

12-18

Glipizide (extended release)

5-20

Glyburide

1.25-20

12-24

Glyburide (micronized)

0.75-12

12-24

Nonsulfonylureas (Meglitinides)

Repaglinide

0.5-16

2-6

Nateglinide

180-360

2-4

GLP-1 agonist

Exenatide

0.01-0.02

4-6

Dipeptidyl Peptidase-4 Inhibitors

Saxagliptin

2.5-5

Sitagliptin

100

12-16

Vildagliptin

50-100

12-24

^aDose may be lower in some patients

K_ATP Channel Modulators: Sulfonylureas

Chemistry. All members of this class of drugs are substituted arylsulfonylureas. They differ by substitutions at the para position on the benzene ring and at one nitrogen residue of the urea moiety. The sulfonylureas are divided into two groups or generations of agents. The first generation sulfonylureas (tolbutamide, tolazamide, and chlorpropamide) are rarely used now in the treatment of type 2 diabetes and are not discussed (information is available in prior editions of this book). The second, more potent generation of hypoglycemic sulfonylureas includes glyburide (glibenclamide; diabeta, others), glipizide (glucotrol, others), and glimepiride (amaryl, others) (Table 43–7). Some are available in an extended-release (glipizide) or a micronized (glyburide) formulation.

Table 43-7Structural Formulas of the Sulfonylureas

Table 43-7 Structural Formulas of the Sulfonylureas

SECOND-GENERATION AGENTS

R₂

Glyburide (Glibenclamide, micronase, diabeta, others)

Glipizide (glucotrol, others)

Gliclazide (diamicron, others; unavailable in the U.S.)

Glimepiride (amaryl)

Mechanism of Action. Sulfonylureas stimulate insulin release by binding to a specific site on the β cell K_ATP channel complex (the sulfonylurea receptor, SUR) and inhibiting its activity. K_ATP channel inhibition causes cell membrane depolarization and the cascade of events leading to insulin secretion (Figure 43–3). The acute administration of sulfonylureas to type 2 diabetes patients increases insulin release from the pancreas. Sulfonylureas may also reduce hepatic clearance of insulin, further increasing plasma insulin levels. In the initial months of sulfonylurea treatment, fasting plasma insulin levels and insulin responses to oral glucose challenges are increased. With chronic administration, circulating insulin levels decline to those that existed before treatment, but despite this reduction in insulin levels, reduced plasma glucose levels are maintained. The explanation for this is not clear, but it may relate to the fact that chronic hyperglycemia per se impairs insulin secretion (glucose toxicity), and with the initial correction of plasma glucose circulating insulin has improved effects on its target tissues. The absence of acute stimulatory effects of sulfonylureas on insulin secretion during chronic treatment is attributed to downregulation of cell surface receptors for sulfonylureas on the pancreatic β cell. If chronic sulfonylurea therapy is discontinued, pancreatic β cell response to acute administration of the drug is restored.

Absorption, Distribution, and Elimination. Although the rates of absorption of the different sulfonylureas vary, all are effectively absorbed from the GI tract. However, food and hyperglycemia can reduce absorption. Sulfonylureas in plasma are largely (90-99%) bound to protein, especially albumin; plasma protein binding is greatest for glyburide. The volumes of distribution of most of the sulfonylureas are ~0.2 L/kg. Although their half-lives are short (3-5 hours), their hypoglycemic effects are evident for 12-24 hours, and they often can be administered once daily. The reason for the discrepancies between their half-lives and duration of action is not clear. The liver metabolizes all sulfonylureas, and the metabolites are excreted in the urine. Thus, sulfonylureas should be administered with caution to patients with either renal or hepatic insufficiency.

Adverse Effects and Drug Interactions. Not unexpectedly, sulfonylureas may cause hypoglycemic reactions, including coma. This is a particular concern in elderly patients with impaired hepatic or renal function who are taking longer-acting sulfonylureas (a major reason first-generation agents are rarely used). Because of the long t_1/2 of some sulfonylureas, it may be necessary to monitor or treat elderly hypoglycemic patients for 24-48 hours with an intravenous glucose infusion in the in-patient setting. Weight gain of 1-3 kg is a common side effect of improving glycemic control with sulfonylurea treatment.

Less frequent side effects of sulfonylureas include nausea and vomiting, cholestatic jaundice, agranulocytosis, aplastic and hemolytic anemias, generalized hypersensitivity reactions, and dermatological reactions. Rarely, patients treated with these drugs develop an alcohol-induced flush similar to that caused by disulfiram or hyponatremia. A long-running debate has centered on whether treatment with sulfonylureas is associated with increased cardiovascular mortality (Bell, 2006). This likely reflects the expression of the sulfonylurea receptor on vascular smooth muscle cells and cardiac myocytes, where activation of the sulfonylurea prevents the beneficial effects of ischemic preconditioning. Glyburide, but not glimepiride, interacts with the sulfonylurea receptor in these non-islet sites and may be associated with increased cardiovascular risk. However, a recent randomized trial of glyburide, metformin, and rosiglitazone found a slightly lower cardiovascular mortality in individuals treated with glyburide (Kahn et al., 2006).

The hypoglycemic effect of sulfonylureas may be enhanced by various mechanisms (decreased hepatic metabolism or renal excretion, displacement from protein-binding sites). For example, some drugs (sulfonamides, clofibrate, and salicylates) displace the sulfonylureas from binding proteins, thereby transiently increasing the concentration of free drug. Ethanol may enhance the action of sulfonylureas and cause hypoglycemia. In addition, hypoglycemia may be more frequent in patients taking a sulfonylurea and one of these agents: androgens, anticoagulants, azole antifungals, chloramphenicol, fenfluramine, fluconazole, gemfibrozil, histamine H₂ antagonists, magnesium salts, methyldopa, monoamine oxidase inhibitors (MAOIs), probenecid, sulfinpyrazone, sulfonamides, tricyclic antidepressants, and urinary acidifiers. Other drugs may decrease the glucose-lowering effect of sulfonylureas by increased hepatic metabolism, increased renal excretion, or inhibiting insulin secretion (β-blockers, Ca²⁺ channel blockers, cholestyramine, diazoxide, estrogens, hydantoins, isoniazid, nicotinic acid, phenothiazines, rifampin, sympathomimetics, thiazide diuretics, and urinary alkalinizers).

Dosage Forms Available. Treatment is initiated at lower end of the dose range and titrated upward based on the patient's glycemic response. Some have a longer duration of action and can be prescribed in a single daily dose (glimepiride), whereas others are formulated as extended-release or micronized formulations to extend their duration of action (Table 43–7). The dose for the extended-release glipizide or micronized glyburide is lower. Glyburide is not recommended when the creatinine clearance is < 50 mL/minute or in elderly individuals because reduced drug and metabolite clearance greatly increases the risk of hypoglycemia. Other sulfonylureas such as glipizide or glimepiride appear safer in elderly individuals with type 2 diabetes.

Therapeutic Uses. Sulfonylureas are used to treat hyperglycemia in type 2 diabetes. Between 50% and 80% of properly selected patients respond to this class of agents. All members of the class appear be equally efficacious. A significant number of patients who respond initially later cease to respond to the sulfonylurea and develop unacceptable hyperglycemia (secondary failure). This may occur as a result of a change in drug metabolism or more likely from a progression of β-cell failure. A recent randomized trial found that the initial improvement in glycemic control was less durable with glyburide monotherapy (compared to metformin or rosiglitazone monotherapy), suggesting that the secondary failure rate is higher with this class of drugs (Kahn et al., 2006). Some individuals with neonatal diabetes or MODY-3 respond to these agents (this is an off-label use). Combinations of insulin and sulfonylureas and other oral agents are discussed later. Contraindications to the use of these drugs include type 1 diabetes, pregnancy, lactation, and, for the older preparations, significant hepatic or renal insufficiency.

K_ATP Channel Modulators: Non-Sulfonylureas

Repaglinide. Repaglinide (prandin) is an oral insulin secretagogue of the meglitinide class (Table 43–6). It is a benzoic acid derivative structurally unrelated to the sulfonylureas, but like sulfonylureas, it stimulates insulin release by closing K_ATP channels in pancreatic β cells. The drug is absorbed rapidly from the GI tract, and peak blood levels are obtained within 1 hour. The t_1/2 is ~1 hour. These features allow for multiple preprandial use as compared with the classical once- or twice-daily dosing of sulfonylureas.

Repaglinide is metabolized primarily by the liver (CYP3A4) to inactive derivatives. Repaglinide should be used cautiously in patients with hepatic insufficiency. Because a small proportion (~10%) is metabolized by the kidney, dosing of the drug in patients with renal insufficiency also should be performed cautiously. As with sulfonylureas, the major side effect of repaglinide is hypoglycemia. Like sulfonylureas, repaglinide is associated with a decline in efficacy (secondary failure) after initially improving glycemic control. Certain drugs may potentiate the action of repaglinide by displacing it from plasma protein binding sites (β-blockers, chloramphenicol, coumarins, MAOIs, nonsteroidal anti-inflammatory drugs [NSAIDs], probenecid, salicylates, and sulfonamide) or altering its metabolism (gemfibrozil, itraconazole, trimethoprim, cyclosporine, simvastatin, clarithromycin).

Nateglinide. Nateglinide (starlix) is an orally effective insulin secretagogue derived from d-phenylalanine. Like sulfonylureas and repaglinide, nateglinide stimulates insulin secretion by blocking K_ATP channels in pancreatic β cells (Rosenstock et al., 2004). Nateglinide promotes a more rapid but less sustained secretion of insulin than other available oral antidiabetic agents. The drug's major therapeutic effect is reducing postprandial glycemic elevations in type 2 diabetes patients. Nateglinide is approved by the FDA for use in type 2 diabetes. It is most effective when administered in a dose of 120 mg, 1-10 minutes before a meal.

Nateglinide is metabolized primarily by hepatic CYPs [2C9, 70%; 3A4, 30%] and thus should be used cautiously in patients with hepatic insufficiency. About 16% of an administered dose is excreted by the kidney as unchanged drug. Dosage adjustment is unnecessary in renal failure. Some drugs reduce the glucose-lowering effect of nateglinide (corticosteroids, rifamycins, sympathomimetics, thiazide diuretics, thyroid products); others (alcohol, NSAIDs, salicylates, MAOIs, and nonselective β blockers) may increase the risk of hypoglycemia with nateglinide. Nateglinide therapy may produce fewer episodes of hypoglycemia than other currently available oral insulin secretagogues including repaglinide. As with sulfonylureas and repaglinide, secondary failure occurs.

A_MPK and PPAR γ Activators

Metformin

Mechanism of Action. Metformin (glucophage, others) is the only member of the biguanide class of oral hypoglycemic drugs available for use today (Bailey and Turner, 1996). Metformin increases the activity of the AMP-dependent protein kinase (AMPK) (Zhou et al., 2001). AMPK is activated by phosphorylation when cellular energy stores are reduced (i.e., lower concentrations of ATP and phosphocreatine). Activated AMPK stimulates fatty acid oxidation, glucose uptake, and nonoxidative metabolism, and it reduces lipogenesis and gluconeogenesis. The net result of these actions is increased glycogen storage in skeletal muscle, lower rates of hepatic glucose production, increased insulin sensitivity, and lower blood glucose levels.

Metformin causes a similar profile of effects and is dependent on AMPK activation (Shaw et al., 2005). Although the molecular mechanism by which metformin activates AMPK is not known, it is thought to be indirect, possibly by reducing intracellular energy stores. Consistent with this, metformin has been shown to inhibit cellular respiration by specific actions on mitochondrial complex I. Metformin has little effect on blood glucose in normoglycemic states and does not affect the release of insulin or other islet hormones and rarely causes hypoglycemia. However, even in persons with only mild hyperglycemia, metformin lowers blood glucose by reducing hepatic glucose production and increasing peripheral glucose uptake. This effect is at least partially mediated by reducing insulin resistance at key target tissues. The hepatic effect is probably the dominant mode of action and involves primarily suppression of gluconeogenesis. Table 43–8 compares metformin and thiazolidinediones (glitazones).

Table 43-8Comparison of Metformin and Thiazolidinediones

Table 43-8 Comparison of Metformin and Thiazolidinediones

PARAMETER

METFORMIN

THIAZOLIDINEDIONES

Molecular target

AMPK

PPARγ

Pharmacologic action

Suppression of HGP

Enhanced insulin sensitivity

Reduction of HbA_1c^a

1.0-1.25

0.5-1.4

Reduction of FFA

Minimal

Moderate

Stimulation of adiponectin

Minimal

Significant

Effect on body weight

Minimal

Increased

Peripheral edema

Minimal

Moderate

Fracture risk

None

Increased

Lactic Acidosis

Rare^b

None

^aMagnitude of absolute reduction dependent on starting A1C value

^bIn renal insufficiency

HGP = hepatic glucose production

Absorption, Distribution, and Elimination. Metformin is absorbed primarily from the small intestine. The drug is stable, does not bind to plasma proteins, and is excreted unchanged in the urine. It has a t_1/2 in the circulation of ~2 hours. The transport of metformin into cells is mediated in part by organic cation transporters (see Chapter 5). Organic cation transporter 1 (OCT 1) is believed to carry the drug into cells such as hepatocytes and myocytes where it is pharmacologically active. Organic cation transporter 2 (OCT 2) is thought to transport metformin into renal tubules for excretion. There is recent evidence suggesting that genetic variation in OCT 1 among humans may affect the response to metformin.

Therapeutic Uses and Dosage. Metformin is currently the most commonly used oral agent to treat type 2 diabetes and is generally accepted as the first-line treatment for this condition. Metformin is effective as monotherapy and in combination with nearly every other therapy for type 2 diabetes, and its utility is supported by data from a large number of clinical trials. Fixed-dose combinations of metformin in conjunction with glipizide, glyburide, pioglitazone, repaglinide, rosiglitazone, and sitagliptin are available. Metformin is available as an immediate-release form, and treatment is best started with low doses and titrated over days to weeks to minimize side effects. The currently recommended dosing is 0.5-1.0 g twice daily, with a maximum dose of 2550 mg; there is no advantage of thrice-daily administration. A sustained-release preparation is available that is effective for once-daily dosing; the maximum dose for this compound is 2 g.

Metformin has superior or equivalent efficacy of glucose lowering compared to other oral agents used to treat diabetes, and reduces diabetes-related complications in patients with type 2 diabetes. Unlike many of the other oral agents, metformin does not typically cause weight gain and in some cases causes weight reduction. Metformin is not effective in the treatment of type 1 diabetes. Several observational studies suggest that diabetic patients treated with metformin may have lower rates of cardiovascular disease and mortality, compared to individuals treated with alternative therapies (Evans et al., 2006; Johnson et al., 2005). The results of the Diabetes Prevention Program indicate that in persons with impaired glucose tolerance, treatment with metformin delays the progression to diabetes. Metformin has been used as a treatment for infertility in women with the polycystic ovarian syndrome. Although not formally approved for this purpose, metformin has demonstrable effects to improve ovulation and menstrual cyclicity and reduce circulating androgens and hirsutism.

Adverse Effects and Drug Interactions. The most common side effects of metformin are gastrointestinal. Approximately 10-25% of patients starting this medication report nausea, indigestion, abdominal cramps or bloating, diarrhea, or some combination of these. Metformin has direct effects on GI function including glucose and bile salt absorption. Use of metformin is associated with 20-30% lower blood levels of vitamin B₁₂ (DeFronzo et al., 1995), probably due to malabsorption, but neurological or hematological consequences of this have not been reported. Most GI adverse effects of metformin abate over time with continued use of the drug, and can be minimized by starting at low doses and gradually titrating to a target dose over several weeks, and by having patients take it with meals.

Like phenformin, metformin has been associated with lactic acidosis. The most concrete evidence of this is from cases of metformin overdose where very high circulating levels of the drug were associated with high plasma lactate and acidemia. However, the estimated incidence of lactic acidosis attributable to metformin use is 3-6 per 100,000 patient-years of treatment (Lalau and Race, 2001; Scarpello and Howlett, 2008), which is comparable to rates in type 2 diabetic patients not using metformin. Moreover, several recent analyses of this association have raised doubts as to whether the association of lactic acidosis with metformin is causal (Kamber et al., 2008). Many cases of lactic acidosis associated with the use of metformin have been reported in patients with concurrent conditions that can cause poor tissue perfusion such as sepsis, myocardial infarction, and congestive heart failure. Renal failure is another common comorbidity reported in patients having lactic acidosis associated with metformin use, and decreased glomerular filtration rates are thought to increase plasma metformin levels by reducing clearance of drug from the circulation. There are no consensus guidelines for renal contraindications for metformin use; because clearance of the drug is not altered significantly until the creatinine clearance drops below 50 mL/minute, metformin is probably safe in patients with this level of renal function (Herrington and Levy, 2008). It is important to assess renal function before starting metformin and to monitor function at least annually. Metformin should be discontinued preemptively in situations where renal function could decline precipitously, such as before radiographic procedures that use contrast dyes and during admission to hospital for severe illness. Metformin should not be used in severe pulmonary disease, decompensated heart failure, severe liver disease, or chronic alcohol abuse.

Cationic drugs that are eliminated by renal tubular secretion have the potential for interaction with metformin by competing for common renal tubular transport systems. Careful patient monitoring and dose adjustment of metformin is recommended in patients who are taking cationic medications such as cimetidine, furosemide, and nifedipine, excreted via the proximal renal tubular secretory system.

Thiazolidinediones

Chemistry and Mechanism of Action. Thiazolidinediones are ligands for the peroxisome proliferation activating receptor γ (PPARγ) receptors, a group of nuclear hormone receptors that are involved in the regulation of genes related to glucose and lipid metabolism (Yki-Jarvinen, 2004). Two thiazolidinediones are currently available to treat patients with type 2 diabetes, rosiglitazone (avandia) and pioglitazone (actos). These compounds are generally similar but have several important differences. Although much is understood about the molecular mechanisms of action of the thiazolidinediones, they have complex effects on a wide variety of tissues, and much remains to be learned about their overall clinical impact and their specific role in therapeutics. Table 43–8 compares metformin and thiazolidinediones.

PPARγ is expressed primarily in adipose tissue with lesser expression in cardiac, skeletal, and smooth muscle cells, islet β cells, macrophages, and vascular endothelial cells. The endogenous ligands for PPARγ include small lipophilic molecules such as oxidized linoleic acid, arachidonic acid and the prostaglandin metabolite 15d-PGJ2; rosiglitazone and pioglitazone are synthetic ligands for PPARγ. Ligand binding to PPARγ causes heterodimer formation with the retinoid X receptor and interaction with PPAR response elements on specific genes, an interaction that is modulated by complex interactions with co-repressors and co-activators. The principal response to PPARγ activation is adipocyte differentiation. Preclinical gain-of-function models show increased numbers of adipocytes and expansion of fat mass; loss-of-function models demonstrate lipodystrophy. Along with adipocyte differentiation, PPARγ activity promotes uptake of circulating fatty acids into fat cells and shifts of lipid stores from extra-adipose to adipose tissue. One consequence of the concerted cellular responses to PPARγ activation is increased tissue sensitivity to insulin, and this is the basis for the pharmacological application of thiazolidinediones to clinical medicine.

Pioglitazone and rosiglitazone are insulin sensitizers and increase insulin-mediated glucose uptake by 30-50% in patients with type 2 diabetes. Although adipose tissue seems to be the primary target for PPARγ agonists, both clinical and preclinical models support a role for skeletal muscle, the major site for insulin-mediated glucose disposal, in the response to thiazolidinediones. It is still not clear whether thiazolidinedione-induced improvement of insulin resistance is due to direct effects on key target tissues (skeletal muscle and liver), indirect effects mediated by secreted products of adipocytes (e.g., adiponectin), or some combination of these. In addition to promoting glucose uptake into muscle and adipose tissue, the thiazolidinediones reduce hepatic glucose production and increase hepatic glucose uptake.

Beyond effects on insulin sensitivity and blood glucose, thiazolidinediones also affect lipid metabolism. Treatment with rosiglitazone or pioglitazone reduces plasma levels of fatty acids by increasing clearance and reducing lipolysis. These drugs also cause a shift of triglyceride stores from nonadipose to adipose tissues, and from visceral to subcutaneous fat depots. In clinical trials, pioglitazone reduces plasma triglycerides by 10-15%, and raises HDL-cholesterol levels. This has been attributed to a dual effect on PPARα as well as PPARγ; rosiglitazone does not interact with PPARα and has minimal effects on plasma triglycerides and cholesterol. Because of these actions on plasma lipids, and beneficial effects on inflammatory markers, coagulation, and endothelial function, thiazolidinediones were predicted to reduce macrovascular complications of diabetes and insulin resistance. However, recently completed randomized clinical trials demonstrated a questionable benefit of pioglitazone (Dormandy et al., 2005), and no effect of rosiglitazone on major events related to atherosclerosis (Home et al., 2009). Because thiazolidinediones reduce hepatic triglyceride, they have been applied to the treatment of nonalcoholic fatty liver disease. Although results of these studies have been encouraging (Kashi et al., 2008), the thiazolidinediones are not yet approved for this use.

Absorption, Distribution, and Elimination. Both agents are absorbed within 2-3 hours, and bioavailability does not seem to be affected by food. The thiazolidinediones are metabolized by the liver and may be administered to patients with renal insufficiency, but should not be used if there is active hepatic disease or significant elevations of serum liver transaminases. The thiazolidinediones are metabolized by hepatic CYPs. CYPs 2C8 and 3A4 metabolize pioglitazone; rosiglitazone is metabolized by CYPs 2C9 and 2C8. Rifampin induces these enzymes and causes a significant decrease in plasma concentrations of rosiglitazone and pioglitazone; gemfibrozil impedes metabolism of the thiazolidinediones and can increase plasma levels by ~2-fold. It may be prudent to reduce the doses of the thiazolidinediones when they are used in conjunction with gemfibrozil. Rosiglitazone and pioglitazone do not seem to significantly affect the pharmacokinetics of other drugs.

Therapeutic Uses and Dosage. Rosiglitazone and pioglitazone are dosed once daily. The starting dose of rosiglitazone is 4 mg and should not exceed 8 mg daily. The starting dose of pioglitazone is 15-30 mg, up to a maximum of 45 mg daily. Thiazolidinediones have proven effects to enhance insulin action on liver, adipose tissue, and skeletal muscle. These composite effects on glucose metabolism confer improvements in glycemic control in persons with type 2 diabetes and cause average reductions in A1C of 0.5-1.4%. Thiazolidinediones require the presence of insulin for pharmacological activity and are not indicated to treat type 1 diabetes. Both pioglitazone and rosiglitazone are effective as monotherapy and as additive therapy to metformin, sulfonylureas, or insulin. The onset of action of thiazolidinediones is relatively slow; maximal effects on glucose homeostasis develop gradually over the course of 1-3 months.

There is evidence in drug-naive diabetic subjects that the glucose-lowering effect of primary treatment with rosiglitazone is more durable than that of metformin or glyburide (Kahn et al., 2006). Treatment with rosiglitazone also reduced progression from prediabetes to type 2 diabetes (Gerstein et al., 2006).

Adverse Effects and Drug Interactions. The most common adverse effects of the thiazolidinediones are weight gain and edema. Edema attributable to thiazolidinedione treatment occurs in up to 10% of patients in clinical trials and has been reported equally with pioglitazone and rosiglitazone. Beyond the effects of edema, treatment with thiazolidinediones causes an increase in body adiposity and an average weight gain of 2-4 kg over the first year of treatment. Importantly, in both preclinical models and in human studies the gain of body weight occurs peripherally, in subcutaneous adipose tissue, and visceral fat changes little or is reduced. The use of insulin with thiazolidinedione treatment roughly doubles the incidence of edema and amount of weight gain, compared with either drug alone.

Macular edema has been reported in patients using both rosiglitazone and pioglitazone (Ryan et al., 2006), usually in association with more general fluid retention. Beyond regular annual retinal exams, diabetic patients taking thiazolidinediones should be observed for visual changes.

Of greatest concern among the adverse effects of thiazolidinediones is the increased incidence of congestive heart failure. This has generally been attributed to the effect of the drugs to cause plasma volume expansion in type 2 diabetic patients who have significantly increased risk for heart failure. There does not appear to be an acute effect of pioglitazone or rosiglitazone to reduce myocardial contractility or ejection fraction. Yet exposure to these drugs over several years in clinical trials has been associated with an increased incidence of heart failure of up to 2-fold (Home et al., 2009). The use of thiazolidinediones in diabetic patients without a history of heart failure, or with compensated heart failure, can be initiated, but monitoring for signs and symptoms of congestive heart failure is important, especially when insulin is also used. Thiazolidinediones should not be used in patients with moderate to severe heart failure, and they should be discontinued in those who develop clinically apparent heart failure while being treated (Nesto et al., 2003).

Recent evidence suggests that rosiglitazone, but not pioglitazone, increases the risk of cardiovascular events (myocardial infarction, stroke). While the degree of the risk remains controversial, an expert panel reviewing this question for the FDA recommended caution in the use of rosiglitazone, but did not recommend its removal from the market (Rosen, 2010). Rather, as of September, 2010, the FDA requires that new prescriptions for rosiglitazone be issued under a risk evaluation and mitigation strategy and be limited to patients whose diabetes could not be adequately controlled by other medications (including pioglitazone). The European Medicines Agency has suspended marketing of rosiglitazone-containing formulations and recommended removal of rosiglitazone and rosiglitazone-containing formulations from the European market.

Evidence from clinical trials indicates that treatment with thiazolidinediones can increase the risk of bone fracture in women (Home et al., 2009; Kahn et al., 2006; Meier et al., 2008). This effect has been proposed to result from increased shunting of mesenchymal stem cells into the adipocyte lineage, and away from osteogenesis, as a result of increased PPARγ activity. Treatment with thiazolidinediones has also been associated with a small but consistent reduction in the hematocrit. This has generally been attributed to hemodilution due to plasma volume expansion, but thiazolidinediones may conceivably shunt erythroid precursors into adipose tissue development.

The first thiazolidinedione released for treatment of diabetic patients, troglitazone, was removed from the market because of rare but severe, and sometimes fatal, hepatic failure. There have been few cases of severe hepatocellular damage attributable to the newer thiazolidinediones, and these have not caused death or required transplantation. In general, pioglitazone and rosiglitazone have been associated with a lowering of transaminases, probably reflective of reductions in hepatic steatosis. It is recommended that thiazolidinediones be withheld from patients with clinically apparent liver disease and that liver function be monitored intermittently during treatment.

GLP-1-Based Agents

Incretins are GI hormones that are released after meals and stimulate insulin secretion. The two best known incretins are GLP-1 and GIP. Although these peptides share many similarities, they differ in that GIP is not effective for stimulating insulin release and lowering blood glucose in persons with type 2 diabetes, whereas GLP-1 is effective. Consequently, the GLP-1 signaling system has been a successful drug target.

Both GLP-1 and glucagon are products derived from preproglucagon, a 180–amino acid precursor with five separately processed domains (Figure 43–9) (Drucker, 2006). An amino-terminal signal peptide is followed by glicentin-related pancreatic peptide, glucagon, GLP-1, and glucagon-like peptide 2 (GLP-2). Processing of the protein is sequential and occurs in a tissue-specific fashion. Pancreatic α-cells cleave proglucagon into glucagon and a large C-terminal peptide that includes both of the GLPs. Intestinal L-cells and specific hindbrain neurons process proglucagon into a large N-terminal peptide that includes glucagon or GLP-1 and GLP-2. GLP-2 impacts the proliferation of epithelial cells lining the GI tract. Teduglutide, a GLP-2 analog, is under development as a treatment for short bowel syndrome and has received orphan drug designation for the treatment of short bowel syndrome from the U.S. FDA and the European Medicines Agency.

Figure 43–9.

Processing of proglucagon to glucagon, GLP1. GLP-2, and GRPP. Proglucagon is synthesized in islet α cells, intestinal enteroendocrine cells (L cells) and a subset of neurons in the hindbrain. In α cells, prohormone processing is primarily by proconvertase 2, releasing glucagon, glicentin-related pancreatic polypeptide (GRPP), and a major proglucagon fragment, containing the two glucagon-like peptides (GLPs). In L cells and neurons, proglucagon cleavage is mostly through proconvertase 1/3, giving the larger C-terminal peptides, glicentin and oxyntomodulin, and the smaller GLP-1 and GLP-2. STN, solitary tract nucleus.

Given intravenously to diabetic subjects in supraphysiologic amounts, GLP-1 stimulates insulin secretion, inhibits glucagon release, delays gastric emptying, reduces food intake, and normalizes fasting and postprandial insulin secretion. The insulinotropic effect of GLP-1 is glucose dependent in that insulin secretion at fasting glucose concentrations, even with high levels of circulating GLP-1, is minimal. The effects of GLP-1 to promote glucose homeostasis and the glucose dependence of these effects are beneficial aspects of this signaling system for treating type 2 diabetes. GLP-1 is rapidly inactivated by the enzyme dipeptidyl peptidase IV (DPP-4), with a plasma t_1/2 of 1-2 minutes; thus, the natural peptide, itself, is not a useful therapeutic agent.

Two broad strategies have been taken to applying GLP-1 to therapeutics, the development of injectable, DPP-4 resistant peptide agonists of the GLP-1 receptor, and the creation of small molecule inhibitors of DPP-4 (Figure 43–10; Table 43–6).

Figure 43–10.

Pharmacological effects of DDP-4 inhibition. DPP-4, an ectoenzyme located on the luminal side of capillary endothelial cells metabolizes the incretins, glucagon-like peptide 1 (GLP-1). and glucose-dependent insulinotropic polypeptide (GIP), by removing the two N-terminal amino acids. The target for DPP-4 cleavage is a proline or alanine residue in the second position of the primary peptide sequence. The truncated metabolites GLP-1[9-36] and GIP[3-42] are the major forms of the incretins in plasma and are inactive as insulin secretagogues. Treatment with a DPP-4 inhibitor increases the concentrations of intact GLP-1 and GIP.

GLP-1 Receptor Agonists. There are currently two GLP-1 receptor agonists that have been approved for treatment diabetic patients in the U.S. and several others in advanced phases of development. Exendin-4 is a naturally occurring reptilian peptide of 39 amino acids with considerable homology to GLP-1. It is a potent GLP-1 receptor agonist that shares many of the physiological and pharmacological effects of GLP-1. It is not metabolized by DPP-4 and so has a plasma t_1/2 of 2-3 hours following subcutaneous injection. Importantly, exendin-4 causes glucose-dependent insulin secretion, delayed gastric emptying, lower glucagon levels, and reduced food intake.

Exenatide (byetta), synthetic exendin-4, is approved for use as monotherapy and as adjunctive therapy for type 2 diabetes patients not achieving glycemic targets with metformin, sulfonylurea, the combination of metformin and sulfonylurea, or thiazolidinedione. In clinical trials, exenatide, alone or in combination with metformin, sulfonylurea, or thiazolidinedione, was associated with improved glycemic control, as reflected in an ~1% decrease in HbA1C, and weight loss that averaged 2.5-4 kg (Amori et al., 2007).

Liraglutide is a second GLP-1 receptor agonist. Structurally, liraglutide is nearly identical to native GLP-1, with Lys³⁴ to Arg substitution and addition of a α-glutamic acid spacer coupled to a C16 fatty acyl group. The fatty acid side chain permits binding to albumin and other plasma proteins and accounts for an extended t_1/2 permitting once a day administration. The pharmacodynamic profile of liraglutide mimics GLP-1 and exenatide, and in clinical trials liraglutide caused both improvement in glycemic control and weight loss. In a single comparative trial, liraglutide reduced A1C~30% more than exenatide (Buse et al., 2009). Liraglutide is indicated for adjunctive therapy in patients not achieving glycemic control with metformin, sulfonylurea, or the combination of metformin/sulfonylurea or metformin/thiazolidinedione.

There are a number of other GLP-1 receptor agonists in development. A sustained-release form of exenatide, exenatide LAR, is being tested as a once-weekly injection. This compound is composed of exenatide in microspheres with the biodegradable polymer D,L lactic-co-glycolic acid. In phase 3 studies, exenatide LAR shares the pharmacodynamic properties of exenatide but may be more potent (Kim et al., 2007). Other long-acting GLP-1 receptor agonists in advanced stages of development include albiglutide, a recombinant protein fusion of GLP-1 and albumin; taspoglutide, a modified GLP-1; lixisenatide, a modified exendin-4 based molecule; and CJC-1134, an exendin-4/albumin conjugate. These agonists all have extended half-lives relative to exenatide and liraglutide and may be suitable for once-weekly injection. The distinguishing feature among the drugs currently available or under investigation is their pharmacokinetics and the exposure time to the GLP-1 receptor.

Mechanism of Action. All GLP-1 receptor agonists share a common mechanism, activation of the GLP-1 receptor. GLP-1 receptors are expressed by β cells, cells in the peripheral and central nervous system, the heart and vasculature, kidney, lung, and GI mucosa. Binding of agonists to the GLP-1 receptor activates the cAMP-PKA pathway and several GEFs (guanine nucleotide exchange factors) (Drucker, 2006). GLP-1 receptor activation also initiates signals through PKC and PI3K, and it alters the activity of several ion channels. In β cells, the end result of these actions is increased insulin biosynthesis and exocytosis in a glucose-dependent manner (Figure 43-3).

Absorption, Distribution, Metabolism, Excretion, and Dosing. Exenatide is given as a subcutaneous injection twice daily, typically before meals. Exenatide is rapidly absorbed, reaches peak concentrations in ~2 hours, undergoes little metabolism in the circulation, and has a volume of distribution of nearly 30 L. Clearance of the drug occurs primarily by glomerular filtration, with tubular proteolysis and minimal reabsorption. Exenatide is marketed as a pen that delivers 5 or 10 μg; dosing is typically started at the lower amount and increased as the response to therapy warrants.

Liraglutide is given as a subcutaneous injection once daily. Peak levels occur in 8-12 hours and the elimination t_1/2 is 12-14 hours. There is little renal or intestinal excretion of liraglutide, and clearance is primarily through the metabolic pathways of large plasma proteins. Liraglutide is supplied in a pen injector that delivers 0.6, 1.2, or 1.8 mg of drug; the low dose is for treatment initiation and generally advanced to the two higher doses based on clinical response.

Adverse Effects and Drug Interactions. The side effects of GLP-1 receptor agonists mimic the pharmacology of native GLP-1. Intravenous or subcutaneous administration of GLP-1 causes nausea and vomiting in a dose-dependent manner; the doses above which GLP-1 causes GI side effects are higher than those needed to regulate blood glucose. Despite this therapeutic window that minimizes adverse effects, nausea, vomiting, and other problems related to GI function are common with these drugs. Based on data from clinical trials, up to 40-50% of subjects given a GLP-1 receptor agonist report nausea at the initiation of therapy. Because the GI side effects of these drugs wane over time, most affected patients are able to continue a course of therapy. The activation of the GLP-1 receptor can delay gastric emptying; thus exenatide and other drugs of this class should be used with caution with other compounds that affect gastric emptying. Moreover, GLP-1 agonists may alter the pharmacokinetics of drugs that require rapid GI absorption, such as oral contraceptives and antibiotics.

In the absence of other diabetes drugs that cause low blood glucose, hypoglycemia associated with GLP-1 agonist treatment is rare. The combination of exenatide or liraglutide with sulfonylurea drugs causes an increased rate of hypoglycemia compared to sulfonylurea treatment alone. Because of the high degree of renal clearance, exenatide should not be given to persons with moderate to severe renal failure (creatinine clearance <30 mL/minute). There is no current recommendation for dose adjustments of other GLP-1 receptor agonists for decreased renal function. Exenatide has also been rarely associated with acute renal failure. Based on surveillance data, there is a possible association of exenatide treatment with pancreatitis, including fatal and nonfatal hemorrhagic or necrotizing pancreatitis. There is currently no mechanism to explain this association, and cases linking exenatide to pancreatitis are rare.

DPP-4 Inhibitors

Mechanism of Action. DPP-4 is a serine protease that is widely distributed throughout the body, expressed as an ectoenzyme on endothelial cells, on the surface of T-lymphocytes, and in a circulating form. DPP-4 cleaves the two N-terminal amino acids from peptides with a proline or alanine in the second position. Although there are many potential substrates for this enzyme, it seems to be especially critical for the inactivation of GLP-1 and GIP (glucose-dependent insulinotropic polypeptide; gastric inhibitory peptide) (Drucker, 2007). DPP-4 inhibitors increase the AUC of GLP-1 and GIP when their secretion is by a meal (Figure 43-10).

Dosage Forms. Several agents provide nearly complete and long-lasting inhibition of DPP-4, thereby increasing the proportion of active GLP-1 from 10-20% of total circulating GLP-1 immunoreactivity to nearly 100%. Two of these, sitagliptin (januvia) and saxagliptin (onglyza), are now available in the U.S.; a third, vildagliptin, is available in the E.U.; a fourth compound, alogliptin, is in advanced stages of clinical trials. Sitagliptin and alogliptin are competitive inhibitors of DPP-4, whereas vildagliptin and saxagliptin bind the enzyme covalently. Despite these differences, all four drugs can be given in doses that lower measurable activity of DPP-4 by >95% for 12 hours. This causes a greater than 2-fold elevation of plasma concentrations of active GIP and GLP-1 and is associated with increased insulin secretion, reduced glucagon levels, and improvements in both fasting and postprandial hyperglycemia. Inhibition of DPP-4 does not appear to have direct effects on insulin sensitivity, gastric motility, or satiety, nor does chronic treatment with DPP-4 inhibitors affect body weight. DPP-4 inhibitors, used as monotherapy in type 2 diabetic patients, reduced HbA_1c levels by an average ~0.8%. These compounds are also effective for chronic glucose control when added to the treatment of diabetic patients receiving metformin, thiazolidinediones, sulfonylureas, and insulin. The effects of DPP-4 inhibitors in combination regimens appear to be additive. The recommended dose of sitagliptin is 100 mg once daily. The recommended dose of saxagliptin is 5 mg once daily.

Absorption, Distribution, Metabolism, and Excretion. DPP-4 inhibitors are absorbed effectively from the small intestine. They circulate in primarily in unbound form and are excreted mostly unchanged in the urine. DPP-4 inhibitors do not bind to albumin, nor do they affect the hepatic cytochrome oxidase system. Both sitagliptin and saxagliptin are excreted renally, and lower doses should be used in patients with reduced renal function. Sitagliptin has minimal metabolism by hepatic microsomal enzymes. Saxagliptin is metabolized by CYP 3A4/5 to an active metabolite. The dose saxagliptin should be lowered to 2.5 mg daily when co-administered with strong CYP3A4 inhibitors (e.g., ketoconazole, atazanavir, clarithromycin, indinavir, itraconazole, nefazodone, nelfinavir, ritonavir, saquinavir, and telithromycin).

Adverse Effects and Drug Interactions. There are no consistent adverse effects that have been noted in clinical trials with any of the DPP-4 inhibitors. With few exceptions the incidence of adverse effects in drug-treated and placebo-treated patients has been similar. DPP-4 is expressed on lymphocytes; in the immunology literature, the enzyme is referred to as CD26. Although there is some evidence of minor effects on in vitro lymphocyte function with DPP-4 inhibitors, there is no evidence from clinical studies of major adverse effects in humans. This area bears scrutiny as more patients are treated with these compounds.

OTHER HYPOGLYCEMIC AGENTS

Alpha Glucosidase Inhibitors

Mechanism of Action. α-Glucosidase inhibitors reduce intestinal absorption of starch, dextrin, and disaccharides by inhibiting the action of α-glucosidase in the intestinal brush border. Inhibition of this enzyme slows the absorption of carbohydrates from the GI tract and blunts the rate of rise of postprandial plasma glucose. These drugs also increase the release of the glucoregulatory hormone GLP-1 into the circulation, which may contribute to their glucose-lowering effects.

Dosage Forms. The drugs in this class are acarbose (precose, others), miglitol (glyset), and voglibose; only acarbose and miglitol are available in the U.S. Dosing of acarbose and miglitol are similar. Both are provided as 25, 50, or 100 mg tablets that are taken before meals. It is recommended that treatment start with lower doses and be titrated as indicated by balancing postprandial glucose, A1C, and GI symptoms. Acarbose and miglitol are most effective when given with a starchy, high-fiber diet with restricted amounts of glucose and sucrose.

Absorption, Distribution, Metabolism, Excretion, and Dosing. Acarbose is minimally absorbed and the small amount of drug reaching the systemic circulation is cleared by the kidney. Miglitol absorption is saturable, with 50-100% of any dose taken into the circulation. Miglitol is cleared almost entirely by the kidney, and dose reductions are recommended for patients with creatinine clearance <30 mL/minute.

Adverse Effects and Drug Interactions. The most prominent adverse effects related to α-glucosidase inhibitors are malabsorption, flatulence, diarrhea, and abdominal bloating. These side effects are dose dependent and related to the mechanism of action of the drugs, with greater amounts of carbohydrate available in the lower intestinal tract for metabolism by bacteria. Mild to moderate elevations of hepatic transaminases have been reported with acarbose, but symptomatic liver disease is very rare. Cutaneous hypersensitivity has been described but is also rare. α-Glucosidase inhibitors do not stimulate insulin release and therefore do not result in hypoglycemia when given alone. Hypoglycemia has been described when α-glucosidase inhibitors are added to insulin or an insulin secretagogue. For hypoglycemia in the setting of α-glucosidase use, glucose rather than sucrose or more complex carbohydrates should be used for treatment. Acarbose can decrease the absorption of digoxin and miglitol can decrease the absorption of propranolol and ranitidine. Alpha glucosidase inhibitors are contraindicated in patients with stage 4 renal failure.

Therapeutic Uses. α-Glucosidase inhibitors are indicated as adjuncts to diet and exercise in type 2 diabetic patients not reaching glycemic targets. They can also be used in combination with other oral antidiabetic agents and/or insulin. In clinical studies α-glucosidase inhibitors reduce A1C by 0.5-0.8%, fasting glucose by ~1 mM and postprandial glucose by 2.0-2.5 mM. These agents do not cause weight gain or have significant effects on plasma lipids. α-Glucosidase inhibitors have been demonstrated to reduce the progression from impaired glucose tolerance to type 2 diabetes.

Pramlintide

Mechanism of Action. Islet amyloid polypeptide (amylin), is a 37–amino acid peptide produced in the pancreatic β cell and secreted with insulin. A synthetic form of amylin with several amino acid modifications to improve bioavailability, pramlintide (symlin), has been developed as a drug for the treatment of diabetes (Hoogwerf et al., 2008). It is thought to mediate its actions through specific binding to the amylin receptor in specific regions of the hindbrain. Activation of the amylin receptor cause reductions in glucagon release, delayed gastric emptying, and satiety.

Absorption, Distribution, Metabolism, Excretion, and Dosing. Pramlintide is administered as a subcutaneous injection prior to meals. Pramlintide is not extensively bound by plasma proteins and has a t_1/2 of 50 minutes. Metabolism and clearance is primarily by the kidney. The doses in patients with type 1 diabetes start at 15 μg and are titrated upward to a maximum of 60 μg; in type 2 diabetes the starting dose is 60 μg and the maximum is 120 μg. Because of differences in the pH of the solution, pramlintide should not be administered in the same syringe as insulin.

Adverse Effects and Drug Interactions. The most common adverse events associated with pramlintide therapy are nausea and hypoglycemia. Although pramlintide alone does not lower blood glucose, addition to insulin at mealtimes has been noted to cause increased rates of hypoglycemia, occasionally severe. It is currently recommended that prandial insulin doses by reduced 30-50% at the time of pramlintide initiation and then retitrated. Because of its effects on GI motility, pramlintide is contraindicated in patients with gastroparesis or other disorders of motility. Because pramlintide can delay gastric emptying, it should be used with caution when combined with other compounds that affect GI motility. Moreover, the pharmacokinetics of drugs that require rapid GI absorption may be altered. Pramlintide is a pregnancyCategory C drug. Pramlintide can be used in persons with moderaterenal disease (creatinine clearance >20 mL/minute).

Therapeutic Uses. Pramlintide is approved for treatment of types 1 and 2 diabetes as an adjunct in patients who take insulin with meals. In clinical trials of type 2 diabetic subjects, pramlintide reduced A1C by 0.5%, lowered insulin requirements, and caused weight loss of 1.0-2.5 kg over 3-6 months. Results in studies of type 1 diabetic subjects were similar, with reductions in A1C of 0.3-0.5%, weight loss of 1-2 kg, and insulin sparing. Pramlintide is now being evaluated as a drug for weight loss in nondiabetic persons.

Bile Acid Binding Resins

Mechanism of Action. Bile acid metabolism is abnormal in patients with type 2 diabetes, and there have been intermittent reports that bile acid binding resins lower plasma glucose in diabetic patients. The mechanism by which bile acid binding and removal from enterohepatic circulation lowers blood glucose has not been established. Bile acid sequestrants could reduce intestinal glucose absorption, although there is no direct evidence of this. Bile acids also act as signaling molecules through nuclear receptors, some of which may act as glucose sensors. The only bile acid sequestrant specifically approved for the treatment of type 2 diabetes is colesevelam (welchol); other bile acid sequestrants likely are also effective.

Absorption, Distribution, Metabolism and Excretion. Colesevelam is provided as a powder for oral solution and as 625-mg tablets; typical usage is three tablets twice a day before lunch and dinner or six tablets prior to the patient's largest meal. The drug is absorbed from the intestinal tract only in trace amounts, so that its distribution is limited to the GI tract.

Adverse Effects and Drug Interactions. The most common side effects of colesevelam are gastrointestinal, with constipation, dyspepsia, abdominal pain, and nausea affecting up to 10% of treated patients. Intestinal obstruction has been reported in persons with previous intestinal surgery or obstruction. Like other bile acid binding resins, colesevelam can increase plasma triglycerides in persons with an inherent tendency to hypertriglyceridemia. In ~400 patient-year exposures in clinical trials, plasma triglycerides in treated patients did not exceed 1000 mg/dL, and no associated pancreatitis was documented; nonetheless, colesevelam should be used cautiously in patients with plasma triglycerides >200 mg/dL. Colesevelam has no effects on systemic drug metabolism but can interfere with the absorption of commonly used drugs like phenytoin, warfarin, verapamil, glyburide, L-thyroxine, and ethinyl estradiol; these medications should be given 4 hours prior to colesevelam. The drug may also interfere with the absorption of fat-soluble vitamins. Colesevelam is a pregnancy Category B drug that has no contraindications in patients with renal or liver disease.

Therapeutic Uses. The bile acid binding resin colesevelam, approved for treatment of hypercholesterolemia, may be used for treatment of type 2 diabetes as an adjunct to diet and exercise (Sonnett et al., 2009). In clinical trials, colesevelam reduced A1C by 0.5% when added to metformin, sulfonylurea, or insulin treatment in type 2 diabetic patients. A new bile acid binding resin, colistilan (not available in the U.S.), also reduced blood glucose in diabetic subjects. This supports the idea that beneficial effects on glucose metabolism are a shared effect among bile acid binding resins.

Miscellaneous Agents

Bromocriptine. A formulation of bromocriptine (cycloset), a dopamine receptor agonist, was approved by the FDA in 2009 for the treatment of type 2 diabetes but is not yet available in the U.S. Bromocriptine is an established treatment for Parkinson disease and hyperprolactinemia (see Chapters 22 and 38). Studies to elucidate the mechanism of bromocriptine to improve blood glucose have not been definitive. In humans, DA agonists do not improve insulin sensitivity, and in cultured islet cells they do not enhance insulin secretion. Effects of bromocriptine on blood glucose may reflect an action in the CNS. The effectiveness of bromocriptine to improve glycemic control is modest (A1C reduction of 0.1-0.4%). The dose range for CYCLOSET is 1.6 mg to 4.8 mg, taken with food in the morning within two hours of awakening. Side effects include nausea, fatigue, dizziness, orthostatic hypotension, vomiting, and headache. The role of this agent in the treatment of type 2 diabetes is uncertain.

COMBINED PHARMACOLOGICAL APPROACHES TO TYPE 2 DIABETES

Progressive Management of Type 2 Diabetes

There are several useful algorithms or flow charts for the treatment of type 2 diabetes (Figure 43–11; Nathan et al., 2009). There are a number of pathways or combination of drugs that are used for treatment of type 2 diabetes if the glucose control does not reach the therapeutic target (Inzucchi,2002; Monami et al., 2008). For example, the addition of a second oral agent, such as an insulin sensitizer, either alone or in combination with an oral insulin secretagogue, also may provide good therapeutic results. Another approach is to introduce basal insulin (often at bedtime) in combination with oral hypoglycemic agents. This combination allows the oral agents to provide postprandial glycemic control while the basal insulin provides the foundation for normalizing fasting or basal glucose levels. Insulin can be combined with various oral antihyperglycemic agents. Combination of glargine with sulfonylureas and/or metformin can reduce both fasting (basal) and postprandial glucose levels. The exact combination of therapies can be guided by an estimation of the β cell secretory reserve in each patient (i.e., a measurement of circulating C peptide level). The progressive insulin deficiency in type 2 diabetes often makes it increasingly difficult to achieve tight glycemic control solely with oral antihyperglycemic agents (Bretzel et al., 2008; Lasserson et al., 2009). Table 43-9 summarizes available pharmacological agents for the treatment of diabetes.

Table 43-9Comparison of Agents Used for Treatment of Diabetes

Table 43-9 Comparison of Agents Used for Treatment of Diabetes

MECHANISM OF ACTION

EXAMPLES

HbA_1C REDUCTION (%)^a

AGENT-SPECIFIC ADVANTAGES

AGENT-SPECIFIC DISADVANTAGES

CONTRAINDICATIONS

Oral

Biguanides^c

↓ Hepatic glucose production

Metformin

1-2

Weight neutral, Do not cause hypoglycemia, inexpensive

Diarrhea, nausea, lactic acidosis

GFR<50 mL/min, CHF, radiographic contrast studies, seriously ill patients, acidosis

α-Glucosidase inhibitors^c

↓ GI glucose absorption

Acarbose, miglitol

0.5-0.8

Reduce postprandial glycemia

GI flatulence, liver function tests

Renal/liver disease

Dipeptidyl peptidase-4 inhibitors^c

Prolong endogenous GLP-1 action

Saxagliptin, sitagliptin, vildagliptin

0.5-1.0

Do not cause hypoglycemia

Reduce dose with renal disease

Insulin secretagogues – Sulfonylureas^c

↑ Insulin secretion

See text and Table 43–8

1-2

Inexpensive

Hypoglycemia, weight gain

Renal/liver disease

Insulin secretagogues – Non-sulfonylureas^c

↑ Insulin secretion

See text and Table 43–8

1-2

Short onset of action, lower postprandial glucose

Hypoglycemia

Renal/liver disease

Thiazolidinediones^c

↓ Insulin resistance, ↑ glucose utilization

Rosiglitazone, pioglitazone

0.5-1.4

Lower insulin requirements

Peripheral edema, CHF, weight gain, fractures, macular edema. Rosiglitazone may increase risk of CV disease

CHF, liver disease

Bile Acid sequestrants^c

Bind bile acids; mechanism of glucose-lowering not known

Colesevelam

0.5

Constipation, dyspepsia, abdominal pain, nausea, ↑ triglycerides, interfere with absorption of other drugs, intestinal obstruction

Parenteral

Insulin

↑ Glucose utilization, ↓ Hepatic glucose production, and other anabolic actions

See text and Table 43–5

Not limited

Known safety profile

Injection, weight gain, hypoglycemia

GLP-1 agonists^c

↑ Insulin, ↓ Glucagon, slow gastric emptying, satiety

Exenatide, liraglutide

0.5-1.0

Weight loss

Injection, nausea, ↑ risk of hypoglycemia with insulin secretagogues, pancreatitis

Renal disease, agents that also slow GI motility, pancreatitis

Amylin agonists^b,c

Slow gastric emptying, ↓ Glucagon

Pramlintide

0.25-0.5

Reduce postprandial glycemia; weight loss

Injection, nausea, ↑ risk of hypoglycemia with insulin

Agents that also slow GI motility

Medical nutrition therapy and physical activity^c

↓ Insulin, resistance, ↑ insulin secretion

Low-calorie, low-fat diet, exercise

1-3

Other health benefits

Compliance difficult, long-term success low?

^aA1C reduction (absolute) depends partly on starting A1C value.

^bUsed in conjunction with insulin for treatment of type 1 diabetes

^cUsed for treatment of type 2 diabetes

Adapted with permission from Harrison's Principles of Internal Medicine, 17e (Fauci AS, Braunwald E, Kasper DL, Hauser SL, Longo DL, Jameson JL, Loscalzo J, eds.) New York: McGraw-Hill, 2008. Copyright © 2008 by The McGraw-Hill Companies, Inc. All rights reserved.

Figure 43–11.

Treatment algorithm for management of type 2 diabetes mellitus. Patients diagnosed with type 2 diabetes, either by fasting glucose, oral glucose tolerance testing, or A1C, should have diabetes education that includes instruction on medical nutrition therapy and physical activity. Most patients newly diagnosed with type 2 diabetes have had subclinical or undiagnosed diabetes for many years previously and should be evaluated for diabetic complications (retinal exam, test for excess protein or albumin excretion in the urine, and clinical evaluation for peripheral neuropathy and vascular insufficiency); common comorbidities (hypertension and dyslipidemia) should be treated. Metformin is the consensus first line of therapy and should be started at the time of diagnosis. Failure to reach the glycemic target, generally a A1C ≤7% within 3-4 months, should prompt the addition of a second oral agent. Reinforce lifestyle interventions at every visit and check A1C every 3 months. Treatment may escalate to metformin plus two oral agents or metformin plus insulin, if necessary.

Emerging Therapies for Diabetes

A number of immunomodulatory approaches are being investigated to prevent or block the autoimmune process that is central to type 1 diabetes. Interventions with immune modifying agents, such as monoclonal antibodies against the lymphocyte receptors CD3, CD20 and CD 25, and immunosuppressant mycophenolate (Chapter 35), are currently ongoing. The use of immunosuppressive agents, such as cyclosporine, glucocorticoids and azathioprine, has been tested in small numbers of patients with new-onset type 1 diabetes. Although some of these studies have demonstrated that immunosuppression can slow β cell loss, this approach has not been widely accepted because of the acute and long-term toxicities of these drugs.

Drugs in Development for Type 2 Diabetes. Inhibitors of the sodium glucose co-transporter 2 (SGLT2) are designed to reduce hyperglycemia by increasing urinary clearance of glucose. SGLT2 is expressed in the proximal renal tubule and functions to reabsorb filtered glucose from the tubule lumen to epithelial cells; glucose subsequently is returned to the circulation through the actions of the glucose transporters 1 and 2. SGLT2 is a low-affinity, high-capacity co-transporter that shuttles sodium and glucose in a 1:1 ratio and accounts for most of the recovery of filtered glucose in the nephron; SGLT1 located in the distal tubule plays a lesser role. Loss-of-function mutations in SGLT2 cause glucosuria. There are several orally available, highly specific SGLT2 inhibitors in clinical development. These compounds cause urinary losses of 50-100 g per day of glucose in the urine, reduce fasting glucose by 0.5-2.0 mM, lower A1C by 0.5-1.0%, and cause modest amounts of weight loss. In clinical trials, SGLT2 inhibitors have been generally well tolerated. There is a mild diuretic effect, rare instances of hypoglycemia, and a slight increase in urinary tract infection, including genital infections. Although there are yet no specific indications with this class of drugs, the novel mechanism, the general action to reduce glycemia in proportion to plasma levels and the blockade of reuptake of renally filtered solute, raise the possibility that SGLT2 inhibitors could have application as adjuncts to a wide range of other diabetes therapies.

A second area of new drug development is activators of glucokinase (GK), the principle hexokinase in islet β cells and hepatocytes. The half-maximal activity of GK is near 8 mM so that the enzyme is highly functional as blood glucose increases from fasting levels of 5 mM to postprandial levels of 7-9 mM. In mouse models engineered to delete GK specifically from the islet or liver, hyperglycemia develops; overexpression of GK leads to improved glucose tolerance. Loss-of-function mutations of GK cause the autosomal dominant form of diabetes, MODY-2. There are now orally available small molecules that are allosteric activators of GK, which lower blood glucose in hyperglycemic animals. Because GK has a relatively low affinity for glucose, the GK activators have muted action at fasting glycemia and a low risk for inducing hypoglycemia. In early phases of clinical trial, GK activators seem to lower blood glucose in diabetic humans.

The enzyme 11β-hydroxysteroid dehydrogenase 1 (11β-HSD1) is expressed in the liver, adipose tissue, and skeletal muscle. 11β-HSD1 converts inactive cortisone into the active glucocorticoid cortisol (Figure 42-6). Cortisol plays an important role in normal counterregulation of insulin action, but excess activity is associated with weight gain, visceral adiposity, insulin resistance, inefficient proinsulin processing, and hyperglycemia. This has led to the strategy of inhibition of 11β-HSD1 to reduce cortisol levels in the visceral and splanchnic beds, potentially increasing insulin sensitivity and glucose metabolism. Potent, orally available, selective inhibitors of 11β-HSD1 that can inhibit the enzyme in liver and adipose tissue are under investigation. Inhibition of 11β-HSD1 improves glycemic control, insulin sensitivity, and total cholesterol in patients with type 2 diabetes.

Several intracellular enzymes serve as nutrient sensors, including SIRT1, AMPK, DGAT1, and ACC2. Compounds that stimulate the activities of these enzymes have the potential to increase glucose utilization, alter fatty acid metabolism, and increase energy expenditure, potentially resulting in improvements of both glucose and lipid metabolism as well as a decrease in body weight. A SIRT1 activator, resveratrol, has been reported to decrease glucose and insulin levels and improve glucose tolerance in subjects with type 2 diabetes. Activators of other peripheral nutrient sensors are in development.

HYPOGLYCEMIA

During prolonged fasting, the blood glucose levels may drop to 3-4 mM in healthy persons with no symptoms of hypoglycemia due to gradual shifts in fuel metabolism to increase the percentage of energy derived from stored lipids. In the absence of prolonged fasting, healthy humans almost never have blood glucose levels <3.5 mM (Cryer, 2008; Cryer et al., 2009). This is due to a highly adapted neuroendocrine counterregulatory system that prevents acute hypoglycemia, a hazardous and potentially lethal situation. The two major clinical scenarios for hypoglycemia are:

treatment of diabetes
inappropriate production of endogenous insulin or an insulin-like substance by a pancreatic islet tumor (insulinoma) or a non-islet tumor

Hypoglycemia in the first scenario can occur either in the fasting or fed state, whereas in the second scenario, hypoglycemia occurs almost exclusively in the fasting or postabsorptive state. Some drugs not used for the treatment of diabetes promote hypoglycemia (Table 43–3; Murad et al., 2009). After a brief discussion of the physiological responses to hypoglycemia, agents used to treat hypoglycemia are discussed.

The most common and serious adverse event related to diabetes therapy is hypoglycemia. Although an adverse reaction to a number of oral therapies, it is most pronounced and serious with insulin therapy. Hypoglycemia may result from an inappropriately large dose, from a mismatch between the time of peak delivery of insulin and food intake, or from superimposition of additional factors that increase sensitivity to insulin (e.g., adrenal or pituitary insufficiency) or that increase insulin-independent glucose uptake (e.g., exercise). The more vigorous the attempt to achieve euglycemia, the more frequent are the episodes of hypoglycemia (Cryer, 2008; Heller, 2008). In the Diabetes Control and Complications Trial, the incidence of severe hypoglycemic reactions was three times higher in the intensive insulin therapy group than in the conventional therapy group. Milder but significant hypoglycemic episodes were much more common than were severe reactions, and their frequency also increased with intensive therapy. Hypoglycemia is the major risk that always must be weighed against benefits of efforts to normalize glucose control.

There is a hierarchy of physiological responses to hypoglycemia. The first response is a reduction of endogenous insulin secretion, which occurs at a plasma glucose level of ~70 mg/dL (3.9 mM); thereafter, the counter-regulatory hormones—epinephrine, glucagon, growth hormone, cortisol, and norepinephrine—are released. Symptoms of hypoglycemia are first discerned at a plasma glucose level of 60-80 mg/dL (3.3-4.4 mM). Sweating, hunger, paresthesias, palpitations, tremor, and anxiety, principally of autonomic origin, usually are seen first. Difficulty in concentrating, confusion, weakness, drowsiness, a feeling of warmth, dizziness, blurred vision, and loss of consciousness (i.e., most importantneuroglycopenic symptoms) usually occur at lower plasma glucose levels than do autonomic symptoms.

Glucagon and epinephrine are the most important counter-regulatory hormones in acute hypoglycemia in newly diagnosed diabetes patients and normal subjects. In patients with type 1 and type 2 diabetes of longer duration, the glucagon secretory response to hypoglycemia becomes deficient, but effective glucose counter-regulation still occurs because epinephrine plays a compensatory role. Diabetic patients thus become dependent on epinephrine for counter-regulation, and if this mechanism becomes deficient, the incidence of severe hypoglycemia increases. This occurs in patients with diabetes of long duration who may also have autonomic neuropathy. The absence of both glucagon and epinephrine can lead to prolonged hypoglycemia, particularly during the night, when some individuals can have extremely low plasma glucose levels for several hours. Severe hypoglycemia can lead to convulsions and coma. In addition to autonomic neuropathy, several related syndromes of defective counter-regulation contribute to the increased incidence of severe hypoglycemia in intensively treated type 1 diabetes patients. These include hypoglycemic unawareness, altered thresholds for release of counter-regulatory hormones, and deficient secretion of counter-regulatory hormones. With the ready availability of home glucose monitoring, hypoglycemia can be documented in most patients who experience suggestive symptoms. Hypoglycemia that occurs during sleep may be difficult to detect but should be suspected from a history of morning headaches, night sweats, or symptoms of hypothermia.

All diabetic patients who receive insulin or oral agents that can cause hypoglycemia should be aware of the symptoms of hypoglycemia, carry some form of easily ingested glucose, and carry an identification card or bracelet containing pertinent medical information. When possible, patients who suspect that they are experiencing hypoglycemia should document the glucose concentration with a measurement. Mild-to-moderate hypoglycemia may be treated simply by ingestion of glucose (15 g of carbohydrate). When hypoglycemia is severe, it should be treated with intravenous glucose or an injection of glucagon (Figure 43–10).

Agents Used to Treat Hypoglycemia

Glucagon is a single-chain polypeptide of 29 amino acids and is produced by recombinant DNA technology. Glucagon interacts with a GPCR on the plasma membrane of target cells that signals through G_s. The primary effects of glucagon on the liver are mediated by cAMP. Glucagon is used to treat severe hypoglycemia, particularly in diabetic patients when the patient cannot safely consume oral glucose and intravenous glucose is not available (Figure 43–12). For hypoglycemic reactions, 1 mg is administered intravenously, intramuscularly, or subcutaneously, usually by the patient's family member, who must be trained to administer this.

Figure 43–12.

Treatment of hypoglycemia with glucose or glucagon. The blood glucose falls following administration of insulin (at time = 0). Following administration of oral glucose (10 or 20 g) or subcutaneous glucagon (at arrow marked "R_x"), the blood glucose rises. The blue band shows the recovery when glucose or glucagon was not administered. (Copyright 1993 American Diabetes Association. From Wiethop and Cryer. Diabetes Care, Vol. 16, 1993; 1131–1136. Reprinted with permission from The American Diabetes Association.)

The intramuscular route is preferred in such an emergency. The hyperglycemic action of glucagon is transient and may be inadequate if hepatic stores of glycogen are depleted. After the initial response to glucagon, patients should be given glucose or urged to eat to prevent recurrent hypoglycemia. Glucagon also is sometimes used to relax the intestinal tract to facilitate radiographic examination of the upper and lower gastrointestinal tract with barium and retrograde ileography and in magnetic resonance imaging of the gastrointestinal tract. Nausea and vomiting are the most frequent adverse effects.

Diazoxide (proglycem) is an antihypertensive, antidiuretic benzothiadiazine derivative with potent hyperglycemic actions when given orally. Hyperglycemia results primarily from inhibition of insulin secretion. Diazoxide interacts with the K_ATP channel on the β cell membrane and either prevents its closing or prolongs the open time; this effect is opposite to that of the sulfonylureas (Figure 43–3). Diazoxide has been used to treat patients with various forms of chronic or recurring hypoglycemia.

The usual oral dose is 3-8 mg/kg per day in adults and children and 8 to 15 mg/kg per day in infants and neonates. The drug can cause nausea and vomiting and thus usually is given in divided doses with meals. Diazoxide circulates largely bound to plasma proteins and has a t_1/2 of ~48 hours. Thus the patient should be maintained at the chosen dosage for several days before evaluating the therapeutic result. Diazoxide has a number of adverse effects, including retention of Na⁺ and fluid, hyperuricemia, hypertrichosis (especially in children), thrombocytopenia, and leucopenia, which sometimes limit its use. Despite these side effects, the drug may be useful in patients with inoperable insulinomas and in children with neonatal hyperinsulinism (leucine sensitivity, islet cell hyperplasia, nesidioblastosis, extrapancreatic malignancy, or islet cell adenoma).

OTHER PANCREATIC ISLET HORMONES

Somatostatin was identified as an inhibitor of GH secretion from the pituitary. Somatostatin (SST) is produced by δ cells of the pancreatic islet, cells of the GI tract, and in the CNS. Somatostatin, produced as a 14–amino acid or a 28–aminoacid peptide molecule, acts through a family of five GPCRs, SSTR_1-5. SST inhibits a wide variety of endocrine and exocrine secretions, including TSH and GH from the pituitary, gastrin, motilin, VIP, glicentin, and insulin, glucagon, and pancreatic polypeptide from the pancreatic islet. The physiological role of somatostatin has not been defined precisely, but its short t_1/2 (3-6 minutes) prevents its use therapeutically. Longer-acting analogs such as octreotide (sandostatin) and lanreotide (somatuline) are useful for treatment of carcinoid tumors, glucagonomas, VIPomas, and acromegaly (Chapter 38).

A depot form of octreotide or lanreotide can be administered intramuscularly every 4 weeks. Octreotide or lanreotide successfully controls excess secretion of growth hormone in most patients, and both have been reported to reduce the size of pituitary tumors in about one-third of cases. Octreotide has been used to treat severe hypoglycemia related to sulfonylurea ingestion (Cryer et al., 2009). Because octreotide also can decrease blood flow to the GI tract, it has been used to treat bleeding esophageal varices, peptic ulcers, and postprandial orthostatic hypotension. Gallbladder abnormalities (stones and biliary sludge) occur frequently with chronic use of the somatostatin analogs, as do GI symptoms. Hypoglycemia, hyperglycemia, hypothyroidism, and goiter have been reported in patients being treated with somatostatin analogs for acromegaly.

BIBLIOGRAPHY