Diabetes mellitus, Notas de estudo de Enfermagem

Baixe Diabetes mellitus e outras Notas de estudo em PDF para Enfermagem, somente na Docsity! Editors: F Belfiore C.E. Mogensen New Concepts in Diabetes and lts Treatment LS Belfiore F, Mogensen CE (eds): New Concepts in Diabetes and Its Treatment. Basel, Karger, 2000, pp 1–2 ............................ Introduction Diabetes mellitus and its complications are clinical conditions of growing importance both from the clinical as well as epidemiological standpoint. The relevance of diabetes at clinical and individual level is given by its life- threatening acute complications and, especially, by its chronic complications affecting several organs and systems, with increased risk for ocular, renal, cardiac, cerebral, nervous and peripheral vascular diseases. The high preva- lence of diabetes in many developed countries or in special ethnic groups, entailing premature disability and mortality, points to its relevance at popula- tion level. It is, therefore, mandatory for both the specialist and the practitioner to be acquainted with the pathophysiological mechanisms, clinical manifesta- tions and, above all, therapy of diabetes mellitus. Recent data showing that control of hyperglycemia may prevent the onset or slow down the progression of complications point to the importance of an appropriate and efficacious treatment. Indeed, the aim of this book is to serve as a tool to provide physicians with the latest views on diagnostic aspects and pathophysiological mechanisms as a premise to go deep into the various facets of the modern management of diabetes. This book begins with introductory chapters on classification and clinical aspects, after which an account is given of insulin secretion as modulated by sulfonylureas and of insulin resistance (in its genetic and acquired components) as modified by diet and the new lipase-inhibitory drug or by metformin (and perhaps troglitazone agents). Insulin therapy of both type 1 and, when re- quired, type 2 diabetes is adequately covered. This is followed by an integrated view of metabolic control, including combined therapy and self-monitoring, in the light of the lesson from DCCT (Diabetes Control and Complications Trial) and UK-PDS (United Kingdom Prospective Diabetes Study). 1 published in 1997, divided into four sections (definition and description of diabetes, classification of diabetes, diagnostic criteria and testing for diabetes), which we summarize in this chapter. Definition and Description of Diabetes mellitus The basis of the metabolic alterations in diabetes is the reduction (to a various degree) of insulin action on insulin-sensitive tissues, due to deficiency of insulin secretion or to insulin resistance or both. The majority of cases of diabetes mellitus falls into two major forms: type 1 and type 2 diabetes. Type 1 Diabetes Immune-Mediated Type 1 Diabetes Type 1 diabetes (previously also named insulin-dependent diabetes mel- litus – IDDM – or juvenile-onset diabetes) is an immune-mediated form of diabetes, which accounts for approximately 5–10% of all diabetics in the West- ern world. It occurs mainly in healthy nonobese children or young adults but may also affect subjects at any age, and results from an absolute deficiency of insulin secretion (evidenced by low or undetectable levels of plasma C- peptide), caused by a cellular-mediated autoimmune destruction of pancreatic b-cells. Although the affected subjects are usually nonobese, the presence of obesity is not incompatible with the diagnosis of type 1 diabetes. The course may be rapid in children and young adults, slower in older patients. Adult patients can retain for some time a residual b-cell function while children and adolescents often show early the effects of severe insulin lack, with a diabetes appearing abruptly over days or weeks and rapidly progressing to acute life- threatening complication (ketoacidotic coma), which may be the first mani- festation of the disease, particularly in presence of precipitating factors such as infections or other stress. Genetic Predisposition. Type 1 diabetes is favored by a not yet fully under- stood genetic predisposition, linked to the HLA system. Pedigree studies of type 1 diabetes families have shown a low prevalence of direct vertical transmission. However, the risk to develop the disease for children who are first-degree relatives of type 1 diabetic patients is between 5 and 10%, the risk being increased when there is haploidentity with the affected sibling and even more when there is HLA identity. It has also been observed that the risk is 5-fold higher for children of a diabetic father compared to children of a diabetic mother (sexual imprinting). Candidate genes for type 1 diabetes have been 4Belfiore/Iannello suggested to occur in chromosomes 2, 6, 11 and 15. However, the major gene seems to be located at the HLA locus in the chromosome 6. Indeed, it is now largely accepted that type 1 diabetes is strongly associated to HLA system, especially with the class II molecules which encode for the D allele. Patients who express the DR3 or DR4 alleles or those who are heterozygous (DR3/ DR4) are especially susceptible to type 1 diabetes. Class I alleles (B8, B15) also seem to be associated to type 1 diabetes as they show linkage disequilibrium, i.e. show nonrandom association with the D alleles. Recently, great importance has been attributed to the DQ locus. It has been shown that DQb1*0301 and DQb1*0302 segregate with DR4 and that DQb1*0201 segregates with DR3. Presence of DQb1*0201 and DQb1*0302 or, especially, the heterozygous state DQb1*0201/0302 entails high risk. On the other hand, DQb1*0502 and DQb1*0602 are associated with the DR2 haplotypes and would be protective. Immunologic Mechanisms. Class II molecules are expressed by macro- phages, endothelial cells and lymphocytes, and are required for the presentation of an antigen to the regulatory T cells, which become activated, thus triggering the immune response. In other words, the favoring HLA haplotypes indicated above permit the interaction of environmental factors (such as certain viral infections or chemical agents) with specific cell membrane components (the HLA molecules), which results in the presentation of the antigen to the regu- latory T lymphocytes, thus triggering an autoimmune mechanism. Several viral infections have been suggested as favoring type 1 diabetes, including Coxsackievirus infections, infectious mononucleosis, mumps, congenital ru- bella, hepatitis and encephalomyocarditis. Some toxins have also been impli- cated. Consumption of cow’s milk during the early life may be an important environmental factor associated with type 1 diabetes development and, because the role of bovine albumin in the induction of b-cell autoimmunity have not been confirmed, b-casein has been suggested as the responsible protein. Virus, toxins, or other factors may directly damage b-cells or favor apoptosis (pro- grammed cell death), or may expose cryptic antigen to the immune system, or may act through molecular mimicry (exogenous molecules similar in amino acid sequence to some endogenous molecules), or they may induce expression of class II molecules in the b-cells (which therefore would become antigen- presenting cells, able to trigger the autoimmune response). An alternative hypothesis which does not rely on exogenous antigen postulates a defective removal of autoreactive T cells, which normally are destroyed in the thymus in the early life. In contrast to the most common form of type 1 diabetes, linked to environmental factors (formerly called type IA), in approximately 10% of all cases of type 1 diabetes (more frequently in females, with HLA- DR3, from 30 to50 years of age), the disease is a primary autoimmune disorder (previously called type IB) and is associated to other endocrine and nonendo- 5Etiological Classification, Pathophysiology and Diagnosis crine autoimmune diseases (Grave’s disease, Hashimoto’s thyroiditis, Addison’s disease, primary gonadal failure, vitiligo, pernicious anemia, connective tissue disease, celiac disease, myasthenia gravis, etc.). This primary autoimmune pathogenesis seems to be confirmed by a persistence of islet cell autoantibodies (ICAs) forever. In 85–90% of patients, diabetes is early associated with one or more serological genetic markers such as ICAs, IAAs (insulin autoantibodies), GAD65 (autoantibodies to glutamic acid decarboxylase) and IA-2 or IA-2b (autoantibodies to tyrosine phosphatase). These autoantibodies disappear over the course of a few years in the majority of patients, and may be the result rather than the cause of the autoimmune process. Clinical Picture. Manifest type 1 diabetes is characterized by symptoms linked to the marked hyperglycemia, such as polyuria (due to the osmotic effect of glucose), polydipsia (to compensate for the water lost with polyuria), polyphagia (to compensate for the energetic substrate glucose lost in the urine), weight loss and fatigue (due to loss of glucose in urine and to dehydration), and blurred vision (due to lens osmotic disturbances). These patients are insulin- dependent for their survival and prone to ketosis; impairment of growth, susceptibility to certain infections, hypertension, lipoprotein metabolism al- terations, periodontal disease and psychosocial dysfunctions are frequent. Idiopathic Type 1 Diabetes The idiopathic diabetes includes some forms of type 1 diabetes (common in individuals of African and Asian origin) due to unknown etiology, with strong genetic inheritance (not HLA-associated), without markers of autoim- munity. There is severe deficit of insulin secretion and tendency to ketoacidosis, with absolute requirement of insulin therapy. Pathophysiology of Type 1 Diabetes The pathophysiological changes occurring in type 1 diabetes as a con- sequence of the severe insulin deficiency may be better understood by comparing the normal picture of the main metabolic pathways, as summarized in figure 1, with the abnormal situation present in type 1 diabetes, outlined in figure 2 (see also chapter III on Insulin Resistance). In type 1 diabetes, the deficit of insulin and the prevalence of counterregulatory hormones, primarily glucagon, leads to the activation of glycogenolysis and gluconeogenesis in liver, with ensuing enhanced hepatic glucose output (HGO). In addition, the deficiency in insulin action results in reduced glucose utilization in peripheral insulin sensitive tissues (primarily muscle) as wellas in activation of lipolysis in the adipose tissue (insulin normally exerts an antilipolytic effect), with enhanced release of FFA. The latter, although they cannot be directly converted into glucose in man, favor gluconeo- genesis in the liver. Combination of enhanced HGO and reduced glucose utiliza- 6Belfiore/Iannello Type 2 Diabetes Type 2 diabetes (previously also named non-insulin-dependent diabetes mellitus – NIDDM – or adult-onset diabetes) occurs in approximately 90–95% of diabetic people in the Western world, resulting from insulin resistance and insufficient compensatory insulin secretion. The disease has an insidious onset and remains asymptomatic and undiagnosed for a long period, even if the moderate hyperglycemia is able to induce severe diabetic late complications. Type 2 diabetes is strongly favored by genetic predisposition. However, although it shows familial aggregation as well as a high concordance (80%) in monozygotic twins, its mode of inheritance is not fully understood. It may well be a polygenic disease. In any case, the risk of offspring and siblings of type 2 diabetic patients to develop the disease is relatively elevated. In addition to the genetic predisposition, favoring environmental factors are involved, such as excessive caloric intake, obesity with increased body fat in the abdominal (visceral) site, sedentary habit, etc. The insulin levels may be normal or even increased (especially in presence of obesity) for a long time, but may decrease in the late stage of the disease. The abnormal carbohydrate metabolism can be early identified measuring fasting glycemia (FPG) or per- forming an oral glucose tolerance test (OGTT). This type of diabetes is nonin- sulin-dependent for survival and is nonketosis prone. Hyperglycemia is usually improved or corrected by diet, weight loss and oral hypoglycemic drugs. In type 2 diabetics an acute life-threatening complication, the nonketotic hyperos- molar coma, can develop whereas ketoacidosis seldom occurs spontaneously, although it may arise during stress, infections or other illnesses. Pathophysiology of Type 2 Diabetes This disease is due to a varying combination of insulin resistance and reduction (especially in the late stage of the disease) in insulin secretion (see chapter II on Insulin Secretion and chapter III on Insulin Resistance). The metabolic alterations are less pronounced than those in type 1 diabetes, out- lined in figure 2 (see also chapter III on Insulin Resistance). Due to insulin resistance (and to enhanced counterregulatory hormones), there is increased HGO (which contributes primarily to fasting hyperglycemia) and reduced peripheral glucose utilization. There is also elevation of plasma FFA (resulting from activation of lipolysis and/or the often enhanced fat mass due to coexisting by peripheral tissues, mainly the muscle (step 8). VLDL and triglycerides are increased because hepatic production (step 28) prevails over peripheral degradation by LPL, primarily at the adipose tissue level (step 29). Ketones are formed at high rate (step 25) because the large amount of Ac-CoA cannot be entirely oxidized to CO2. 9Etiological Classification, Pathophysiology and Diagnosis obesity), which in turn contributes to insulin resistance through the mechanism of the glucose-FFA cycle. As mentioned above (under Type 1 Diabetes), hyper- glycemia itself favors glucose utilization (glucose effectiveness). This mecha- nism may be impaired in type 2 diabetes, i.e. ‘glucose resistance’ may be present. It has been observed that in obesity and type 2 diabetes (as well as in acromegaly and Cushing’s disease), in the postabsorptive period, noninsulin- mediated glucose uptake is a major determinant of glucose disposal and is similar in the different pathologies studied. On the other hand, although absolute rates of basal insulin-mediated glucose uptake are reduced in insulin- resistant states, they do not achieve statistical value compared with control subjects because of compensatory hyperinsulinemia. Other Specific Types of Diabetes Various, less common, types of diabetes are known to occur, in which the secretory defect is based upon different mechanisms. Genetic Defects of b-Cell Function The maturity-onset diabetes of the young (MODY) is a genetically hetero- geneous monogenic form of noninsulin-dependent diabetes, characterized by early onset, usually before 25 years of age and often in adolescence or child- hood, and by autosomal dominant inheritance. There is no HLA association nor evidence of cell-mediated autoimmunity. It has been estimated that 2–5% of patients with type 2 diabetes may have this form of diabetes mellitus. However, the frequency of MODY is probably underestimated. Clinical studies have shown that prediabetic MODY subjects have normal insulin sensitivity but suffer from a defect in glucose-stimulated insulin secretion, suggesting that pancreatic b-cell dysfunction, rather than insulin resistance, is the primary defect in this disorder. To date, three MODY genes have been identified. MODY-1. Studies in an affected family showed that the gene responsible for MODY-1 is tightly linked to the adenosine deaminase gene on chromosome 20q. Further research has shown that responsible for MODY-1 is a mutation in the gene-encoding hepatocyte nuclear factor (HNF)-4a, a member of the steroid/thyroid hormone receptor superfamily and an upstream regulator of HNF-1a expression. MODY-2. This form is due to mutations in glucokinase (GK – see chapter II for the functional meaning of GK in b-cells) and is associated with defects in insulin secretion, reduction in hepatic glycogen synthesis and in the net accumulation of hepatic glycogen as well as increased hepatic gluconeogenesis following meals, resulting in impaired glucose tolerance or diabetes mellitus 10Belfiore/Iannello characterized by mild chronic hyperglycemia. The hyperglycemia due to GK deficiency is often mild (fewer than 50% of subjects have overt diabetes) and is evident during the early years of life. Despite the long duration of hyperglycemia, GK-deficient subjects have a low prevalence of micro- and macrovascular complications of diabetes. Obesity, arterial hypertension and dyslipidemia are also uncommon in this form of diabetes. MODY-3. In several families, this form of MODY was found to be linked with microsatellite markers on chromosome 12q. The disease was estimated to be linked to this chromosome region in approximately 50% of families in a heter- ogeneity analysis. It is the most common form of MODY. Affected patients ex- hibit major hyperglycemia with a severe insulin secretory defect, suggesting that the causal gene is implicated in pancreatic b-cell function. MODY-3 was further shown to be due to mutations in the gene-encoding HNF-1a (which is encoded by the gene TCF1). HNF-1a is a transcription factor that helps in the tissue- specific regulation of the expression of several liver genes and also functions as a weak transactivator of the rat insulin-I gene. Familial Hyperinsulinemia. The high-affinity sulfonylurea receptor, a novel member of the ATP-binding cassette superfamily, is one component of the ATP-sensitive K+ channel. The protein is critical for regulation of insulin secretion from pancreatic b-cells, and mutations in the receptor (or in the KATP channels) have been linked to familial hyperinsulinemia, a disorder character- ized by unregulated insulin release despite severe hypoglycemia. Other forms may be due to mutation in the GK gene, leading to a hyperresponsive enzyme. Other. In addition, a diabetes type associated with deafness may be linked to point mutations in mitochondrial DNA, and still other forms with less clearly defined defects are known to occur. In about 50% of cases of MODY, the genetic background is uncertain. It should be stressed that the role of the above genes (responsible for b-cell dysfunction) in the susceptibility to the more common late-onset form of type 2 diabetes remains uncertain. Genetic studies seem to exclude any function as major susceptibility genes, although they might play a minor role in a polygenic context or a major role in particular populations. Rare Genetic Defects of Insulin Action These are a heterogeneous group of rare conditions which includes: (a) syn- dromes associated with acanthosis nigricans, which is a brown to almost black hyperpigmentation of the skin, most often located in the neck, axilla, groin or other areas, less rare in Blacks or in subjects of Hispanic origin. The affected patients show high insulin levels. Some cases are due to mutation in the insulin receptor resulting in diminished tyrosine-kinase activity (type A syndrome). Others are due to antibodies to the insulin receptors which prevent insulin 11Etiological Classification, Pathophysiology and Diagnosis ‘NIDDM’ (which are confusing as they classified the patient according to treat- ment rather than etiology). (b) Preservation of the terms ‘type 1’ or ‘type 2’ diabetes (with Arabic numerals) and elimination of the confusing terms ‘type I’ or ‘type II’ diabetes (with Roman numerals); patients with no evidence of autoimmunity are classified as being affected by type 1 idiopathic diabetes. (c) Type 1 diabetes does not include those forms of b-cell destruction due to nonautoimmune-specific causes. (d) Type 2 diabetes includes the most common form characterized by insulin resistance and insulin secretory defect. (e) The class previously named malnutrition-related diabetes mellitus has been elimi- nated. (f ) The IGT stage has been retained, and the stage of IFG was added. (g) GDM, as defined by WHO and NDDG, was retained. Diagnostic Criteria for Diabetes mellitus A precocious diagnosis of diabetes is important to prevent or attenuate late diabetic complication, and depends upon the adequate use and interpreta- tion of laboratory tests (especially in absence of specific symptoms). Many different diagnostic schemes have been in use. Recently, on the basis of the available data, the diagnostic criteria previously recommended by NDDG or WHO were modified. According to the revised criteria by the Expert Commit- tee [1997], the ‘normal values’ and the ‘diagnostic values’ for diabetes (which do not coincide with the goals of therapy) are as follows (values given in the text refer to venous plasma glucose which is the preferred measurement; equivalents for whole blood and capillary glucose estimations, according to the IDF guidelines [1999] to type 2 diabetes, are indicated in footnotes). Normal Values. The upper limit of normal venous plasma values has been set at 110 mg/dl (6.1 mmol/l) for FPG and at 140 mg/dl (7.8 mmol/l) for the 2-hour value after glucose load (OGTT). Diagnostic Values. (a) FPG q126 mg/dl (or 7.0 mmol/l)1 after a fasting of at least 8 h, confirmed on a subsequent day, to rule out a labeling or technical error; (b) 2-hour value during OGTT q200 mg/dl (or q11.1 mmol/l)2, con- firmed in a repeated test to make the final diagnosis; (c) symptoms of diabetes and a casual value q200 mg/dl (or 11.1 mmol/l) at any time of day. For epidemiological studies, diabetes prevalence and incidence should be estimated by a FPG q126 mg/dl. The value of FPG was changed from the 1 Same value for capillary plasma glucose; q110 mg/dl (?6.0 mmol/l) for venous or capillary whole blood glucose. 2 q220 mg/dl (q12.2 mmol/l) for capillary plasma glucose; q180 mg/dl (q10.0 mmol/l) for venous whole blood glucose; q200 mg/dl (q11.0 mmol/l) for capillary whole blood glucose. 14Belfiore/Iannello previous value (q140 mg/dl) to current value (q126 mg/dl), because (1) the cutpoint of FPG q140 mg/dl defines a greater degree of hyperglycemia than did the cutpoint of the 2-hour value q200 mg/dl, and (2) this degree of hyperglycemia usually reflects a serious abnormality associated with serious chronic diabetic complications. The 2-hour value q200 mg/dl has been re- tained for the diagnosis of diabetes because it was well accepted, and enormous clinical and epidemiological data are based on this cutpoint value. The criteria for diagnosis of diabetes in an asymptomatic child should be stricter than those for the adults to avoid overdiagnosis of diabetes, and it should be considered that normal children commonly present OGTT values lower than adults. The diagnostic values for GDM as proposed by O’Sullivan and Mahan [1993], revised by NDDG and adopted by ADA and the American College of Obstetricians and Gynecologists (ACOG), are set lower than those for nonpregnant adults. A screening test is indicated between 24 and 28 weeks of gestation in asymptomatic female patients at risk, and a value 1 h after a 50 g of glucose load q140 mg/dl (or 7.8 mmol/l) can identify the individuals at risk for GDM in whom a full diagnostic 3-hour OGTT with 100 g of glucose should be performed. GDM occurs with an FPG q105 mg/dl (or 5.8 mmol/l) and a 2-hour value during OGTT q165 mg/dl (or 9.2 mmol/l). An intermediate metabolic state was introduced, which is characterized by glucose levels above those considered as normal but below those accepted for the diagnosis of diabetes mellitus. Referring to the fasting state, this condition was named impaired fasting glycemia or IFG (FPG q110 but p126 mg/dl or q6.0 but p7.0 mmol/l )3. Referring to the postload state, it was named impaired glucose tolerance or IGT (2-hour postload value in OGTT q140 mg/dl but p200 mg/dl or q7.8 but p11.1 mmol/l)4, without spontaneous hyperglycemia). IFG or IGT are not clinical entities but rather risk factors for future type 2 diabetes and cardiovascular disease, being associated with the metabolic syndrome or insulin resistance syndrome, charac- terized by abdominal or visceral obesity, hypertension, dyslipidemia (hypertri- glyceridemia and low HDL value) and hyperuricemia. Conversion of IGT to type 2 diabetes takes years or decades and occurs in about 10–50% of IGT patients. Thus, IGT may not progress to overt diabetes and may revert to normoglycemia, especially in obese patients after dietary treatment and weight reduction. 3 Same value for capillary plasma glucose; q100 but p110 mg/dl (q5.5 but p6.0 mmol/l) for venous or capillary whole blood glucose. 4 q160 but p220 mg/dl (q8.9 but p12.2 mmol/l) for capillary plasma glucose; q120 but p180 mg/dl (q6.7 but p10.0 mmol/l) for venous whole blood glucose; q140 but p200 mg/dl (q7.8 but p11.1 mmol/l) for capillary whole blood glucose. 15Etiological Classification, Pathophysiology and Diagnosis Table 2. Subjects in whom OGTT should be performed First-degree relative of type 2 diabetic patients (especially if monozygotic twin of a diabetic patient or offspring of two diabetic parents) Subjects with abnormal or borderline glycemic values (FPG q110 mg/dl but p126 mg/dl) during screening test for diabetes Pregnant women with suspected GDM Obese subjects (especially when a family history of diabetes is present) Individuals with a family history of MODY Members of racial or ethnic groups with high prevalence of diabetes (American Indians or Pacific Islanders, African-Americans, Hispanics, etc.) Patients with unexplained neuropathy or coronary disease or peripheral vascular disease or retinopathy or nephropathy (especially under 50 years of age) Patients with hyperglycemia or glycosuria found during acute illness, stress situations, surgical procedures, steroid administration, etc. Oral Glucose Tolerance Test The OGTT is not recommended for routine clinical use (being a nonspe- cific test) and should be standardized for both procedure and interpretation, while the use of FPG is encouraged as a simple, convenient, accurate, acceptable to patients and low cost test for diagnosing diabetes. FPG and 2-hour OGTT values are equivalent for the diagnosis of diabetes (even if not perfectly corre- lated with each other), and actually the FPG alone is preferable for its better reproducibility (6% variation) whereas OGTT, repeated in adults during a 2- to 6-week interval, presents an intraindividual coefficient of variation of 17% for the 2-hour value. OGTT remains, however, the most sensitive and practical test for the early recognition of asymptomatic diabetes without high FPG value, and it is an invaluable tool in research studies. If the OGTT is used, the test procedures recommended are that of WHO. The indications of OGTT are outlined in table 2. The following variables may affect the OGTT results: Technical Variables. Venous versus capillary blood: In adults venous blood from an antecubital vein is usually employed, obtained with minimum stasis. In the capillary blood, glucose approximates that of arterial blood, and is higher than in venous blood by 2–3 mg/dl in the fasting state and by 20–70 mg/dl during OGTT. Plasma or serum versus whole blood: Plasma or serum is generally em- ployed, providing more stable values. In these materials glucose concentration is 15% higher than in whole blood. The blood sample should be immediately refrigerated to prevent glycolysis of glucose by blood cells (fluoride cannot be 16Belfiore/Iannello or delay the clinical onset of disease (when a positive autoantibody test is obtained); (c) the cost-effectiveness of the screening is questionable. The au- toantibody tests, however, may be useful to detect which newly diagnosed patients have immune-mediated type 1 diabetes. Type 2 Diabetes. Type 2 diabetes is commonly undiagnosed in about 50% of affected subjects. On the other hand, retinopathy may develop early, even 7 years before the diagnosis of overt diabetes. Thus, the unapparent hyperglyce- mia can cause microvascular complications and favor macrovascular disease. Therefore, the undiagnosed diabetes is a serious problem. Early detection and treatment are indispensable to reduce the late complications of type 2 diabetes. Thus, testing for diabetes (especially with FPG) should be recommended in the clinical setting and in high-risk subjects. In asymptomatic and undiagnosed individuals, testing for type 2 diabetes by FPG should be performed in: (a) all individuals at age 45 and above, repeated at 3-year intervals if results are normal; (b) individuals at younger age if at risk (obese subjects, first-degree relatives of diabetic patients, compo- nents of high-risk ethnic populations, women with GDM, mothers of obese baby ?9 lb or 4 kg, etc.); (c) hypertensive subjects with low HDL cholesterol (p35 mg/dl) or high triglycerides (q250 mg/dl); (d) individuals with IGT or IFG on previous testing. Suggested Reading Expert Committee on the Diagnosis and Classification of Diabetes mellitus: Report of the Expert Committee on the Diagnosis and Classification of Diabetes mellitus. Diabetes Care 1997;20:1183–1197. Fajans SS: Classification and diagnosis of diabetes; in Rifkin H, Porte D (eds): Diabetes mellitus. Theory and Practice, ed 4. New York, Elsevier, 1990, pp 346–356. International Diabetes Federation (IDF), 1998–1999 European Diabetes Police Group: A Desktop Guide to Type 2 (Non-Insulin-Dependent) Diabetes mellitus. Brussels, IDF, 1999. National Diabetes Data Group: Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 1979;28:1039–1057. O’Sullivan JB: Diabetes mellitus after GDM. Diabetes 1993;40(suppl):131–135. Velho G, Blanche H, Vaxillaire M, et al: Identification of 14 new glucokinase mutations and description of the clinical profile of 42 MODY-2 families. Diabetologia 1997;40:217–224. World Health Organization: Diabetes mellitus: Report of a WHO Study Group. Tech Rep Ser No 727. Geneva, WHO, 1985. Yamagata K, Furuta H, Oda N, et al: Mutations in the hepatocyte nuclear factor-4a gene in maturity-onset diabetes of the young (MODY-1). Nature 1996;384:458–460. Yamagata K, Oda N, Kaisaki PJ, et al: Mutations in the hepatocyte nuclear factor-1a gene in maturity-onset diabetes of the young (MODY-3). Nature 1996;384:455–458. F. Belfiore, Institute of Internal Medicine, University of Catania, Ospedale Garibaldi, I–95123 Catania (Italy) Tel. +39 095 330981, Fax +39 095 310899, E-Mail francesco.belfiore@iol.it 19Etiological Classification, Pathophysiology and Diagnosis Chapter II Belfiore F, Mogensen CE (eds): New Concepts in Diabetes and Its Treatment. Basel, Karger, 2000, pp 20–37 ............................ Insulin Secretion and Its Pharmacological Stimulation F. Belfiore, S. Iannello Institute of Internal Medicine, University of Catania, Ospedale Garibaldi, Catania, Italy Insulin Secretion Introduction Pancreatic b-cells synthesize a large polypeptide chain, the proinsulin, which is then cleaved into the so-called connecting peptide (C-peptide) and the insulin molecule, composed of two peptide chains containing 51 amino acid residues. Both insulin and C-peptide are packaged in the secretory gran- ules. During the secretory process, the granule content is discharged outside the b-cell through a process of exocytosis, leading to the release of insulin and C-peptide in equimolar amounts, together with small quantities of uncleaved proinsulin. In contrast to insulin, C-peptide is not taken up by the liver (and the other insulin-sensitive tissues), and therefore its plasma level is a good index of insulin secretion. Regulation of Insulin Secretion by Substrates Glucose Glucose is the main physiological regulator of insulin secretion. In vitro, prolonged stimulation with glucose (or sulfonylureas) induces a biphasic insu- lin secretory response by pancreatic islets characterized by an initial rapid first phase lasting about 5 min, during which about 2–3% of the insulin content of pancreas is released, followed by a slower second phase in insulin secretion, which results in the liberation of up to 20% of total pancreatic content during 20 a period of 60 min of glucose perfusion. A similar biphasic pattern of secretory response to glucose has also been reported in vivo in man with the hyper- glycemic clamp technique. The two secretory phases, however, are not apparent after a carbohydrate-rich meal because the elevation in blood glucose is not rapid enough. Nevertheless, an efficient initial insulin secretory response (dependent upon the b-cell sensitivity to glucose elevation) is required for an optimal glucose control and for avoiding an excessive secretion during the second phase, which entails the risk of late hypoglycemia (reactive hypogly- cemia). Glucose, besides its direct stimulation of insulin release, also potentiates the secretory response to nonglucose stimuli, which may play a role during the absorption of mixed meals. In addition, glucose exerts a priming effect of b- cells, as a previous exposure of b-cells to glucose causes an enhanced secretory response to a subsequent stimulation with glucose (or even with nonglucose stimuli), as if the b-cell has memory of the previous glucose exposure. Chronic exposure to glucose, however, induces desensitization of b-cells, which does not seem to be due to a reduced content or synthesis of insulin. This is relevant to the condition of persistent hyperglycemia occurring in the diabetic state. With regard to insulin secretion, three concepts should be distinguished: the set point for blood glucose, the b-cell threshold for glucose, and the b- cell glucose sensor. The set point entails the concept that there is a control system that ‘sets’ the level of glucose at a given value, which in man is fixed to about 5 mmol/l glucose. The set point is the result of the activity of b-cells as well as of a-cells and d-cells. The glucose threshold for both b-cells and a-cells is between 5 and 6 mmol/l: when glucose rises above this level, the insulin-secreting b-cells are turned on whereas the glucagon-secreting a-cells are turned off, and vice versa. Glucose threshold increases during starvation, when b-cells are blind to even relatively high glucose levels, and returns to normal upon refeeding. In order to be able to respond to increase in glucose concentration above the threshold value, b-cells must be equipped with a glucose sensor, which has been identified in the glucose-phosphorylating enzyme glucokinase (GK). This enzyme, long known to be present in the liver, has been shown to occur also in the b-cells (the liver and the b-cell enzymes differ at genetic level). GK differs from the ubiquitous enzyme hexokinase (which catalyzes the same reaction as GK, i.e. glucose phosphorylation), in that hexokinase has a high affinity (or a low Km) for glucose and therefore works at the maximum activity at very low, under- physiological glucose concentration, whereas GK has a low affinity (or a high Km) for glucose, which entails that its activity increases with increasing glucose concentration. Its presence in the liver allows this organ to take up glucose when glycemia in the portal vein increases (such as during the absorption 21Insulin Secretion and Its Pharmacological Stimulation Regulation of Insulin Secretion by Hormones and Neurotransmitters Acetylcholine, produced by parasympathetic activity, stimulates insulin secretion through muscarinic receptors (which can be blocked by atropine), probably by enhancing DAG (diacylglycerol) and IP3 (inositol-3-P) formation (fig. 1). Parasympathetic stimulation may occur during the early (cephalic and intestinal) phase of insulin secretion following a meal as well as during hypoglycemic episodes. In the latter instance, however, hypoglycemia limits the parasympathetic effect on insulin secretion, because this effect is glucose- dependent. The parasympathetic innervation of the pancreas may also trigger the release of vasoactive intestinal polypeptide (VIP), which stimulates the secretion of insulin (and glucagon) while increasing the blood flow to the pancreas and the external pancreatic secretion. Norepinephrine (released upon sympathetic stimulation) and epinephrine (produced by adrenal medulla) exert both an inhibitory effect, through the a- adrenergic receptors (fig. 1), and a stimulatory effect, through the b-adrenergic receptors, the overall effect being an inhibition of glucose-stimulated insulin release and a little effect in the basal state. Sympathetic nerve activity may also release other neurotransmitters, such as galanin, which would inhibit both basal and stimulated insulin secretion. Gastrointestinal hormones (or gut hormones) contribute to the overall insulin secretion, as shown by the higher insulin secretion after glucose given per os compared to intravenous glucose. For this action, they are also called incretins. They include: the gastric inhibitory polypeptide (GIP), secreted by the endocrine cells of duodenum and jejunum; cholecystokinin (CCK), both the long (CCK-33) and the short (CCK-8) peptide chain, released by duo- denum and proximal part of jejunum after ingestion of fats and proteins; the glucagon-like peptide-1 (7–36) amide, or GLP-1 (7–36), formed from GLP-1 (the precursor proglucagon, produced by the L-cells in the distal part of small intestine, is processed by tissue-specific proteolysis to produce glucagon in pancreatic a-cells and GLP-1 in the intestine), is released after carbohydrate- rich meals (fig. 1); the neuropeptide Y (NPY), a neurotransmitter present in both the central nervous system and the enteric nervous system which produces stimulation of food intake (and of resting metabolic rate), while probably acting as an incretin to enhance insulin release. The counterregulatory hormones (or stress hormones) also affect insulin secretion. Glucagon is a potent stimulus for the islet b-cell (fig. 1), and intrave- nous bolus injection of 1 mg glucagon has been widely used to assess endogen- ous insulin secretion for clinical or research purposes. Glucagon stimulates insulin release mainly through glucagon receptors but not GLP-1 receptors on islet b-cells. On the other hand, insulin may affect glucagon secretion 24Belfiore/Iannello because capillaries go from the central part of the islets, where insulin is mainly produced, to the periphery of the islets where glucagon-producing a-cells are mainly located. The other stress hormones affect insulin secretion through a generally inhibitory action. The effect of epinephrine has been mentioned above. Cortisol and growth hormone (GH) are thought to play a role during prolonged stress periods. Leptin may potentiate glucose-induced insulin secretion by a mechanism involving cAMP or phospholipase C/protein kinase C activation. It also in- hibits NPY release. In contrast to early studies, recent data indicate that amylin is a third active pancreatic islet hormone that works with insulin and glucagon to maintain glucose homeostasis. It would regulate glucose inflow to the circulation by influencing the rate of gastric emptying and would also inhibit hepatic glucose production in the postprandial period. Assessment of Insulin Secretion Fasting Insulin Level. The fasting insulin level (normally between 5 and 15 lU/ml) may reflect the insulin secretory capacity. It may be very low (=5 lU/ml) in subjects with high insulin sensitivity (lean and/or trained sub- jects) and elevated (?15 lU/ml) in insulin-resistant subjects. It should be pointed out that an apparent normal insulin level in insulin-resistant diabetic subjects indicates decreased secretory capacity, since an equal level of glucose in a ‘normal’ subject would be associated with a higher insulin level which would promptly normalize glucose. It should be pointed out that the insulin values usually referred to are those obtained with the commonly used radioimmunoassay method, which yields the total insulin levels, whereas more sophisticated methods are available that allow to distinguish the true insulin from the proinsulin. True insulin may be lower than total insulin by 15–20%. Acute Insulin Response to Glucose (AIRG). AIRG following glucose given as intravenous bolus consists of a rapid increase in insulin level which returns towards normal within 10 min. The magnitude of AIRG is not affected by the preexisting glucose level, which makes this test feasible even in diabetic patients. AIRG is often absent in patients with type 2 diabetes whereas it is enhanced in insulin-resistant obese subjects. Acute Insulin Response to Non-Glucose Stimuli (AIRNG). AIRNG includes response to amino acids, neurotransmitters and hormones. AIRNG obtained with arginine increases with the increase in the preexisting glucose level. By plotting the AIRNG values against those of glycemia, a ‘curve’ is obtained which reflects the correlation between these two variables. From the analysis 25Insulin Secretion and Its Pharmacological Stimulation of this curve, it is possible to deduce the insulin secretory capacity of the pancreas, the maximal acute insulin response to nonglucose stimuli or AIRMAX , and the b-cell sensitivity to the potentiation effect of glucose. The AIRMAX (which indicates the maximal secretory response) is reduced in type 2 diabetes and may increase in insulin-resistant hyperinsulinemic subjects. The b-cell sensitivity to the potentiation effect of glucose is little changed in type 2 patients, suggesting preserved b-cell sensitivity in these patients. Insulin Secretion in Type 2 Diabetes Both impaired insulin action (insulin resistance) and reduced insulin secre- tion (insulin deficiency) may contribute to the development of type 2 diabetes. It is now accepted that in type 2 diabetes the situation may range from predominantly insulin resistance with relative insulin deficiency to a predomi- nantly secretory defect with insulin resistance. It is noteworthy that recent data would suggest that the hyperinsulinemia of insulin resistance may result from an increase in insulin secretion secondary to increased b-cell sensitivity to glucose, as well as a decrease in insulin clearance. In type 2 diabetes the b-cell mass is reduced by about 50%, which is known from experimental pancreatectomy to be not enough to cause fasting hyperglycemia. Therefore, in most type 2 patients a functional defect in b-cells may occur, leading to insulin secretory defect. This is confirmed by the almost absent acute insulin response to glucose (AIRG), diminished maximal acute insulin response to nonglucose stimuli (AIRMAX), decreased insulin secretory capacity, with nor- mal b-cell sensitivity to the potentiation effect of glucose. Type 2 diabetes subjects have their own ‘set’ of fasting plasma glucose which is more or less increased compared to normal. This is due to reduced insulin secretion into portal vein which is unable to completely suppress hepatic glucose production, resulting in hyperglycemia. The latter, on the other hand, is somewhat ‘useful’ in that it forces the hypofunctioning b-cell to secrete more insulin. The apparently ‘normal’ fasting insulin in type 2 diabetes in the presence of fasting hyperglycemia should indeed be considered as reduced. In fact, restoring normal glucose levels in mild diabetes by an insulin infusion reduces the endogenous insulin concentration to subnormal values. The same reasoning apply for the insulin response to oral glucose, i.e. to the insulin curve during OGTT. The insulin response (area under the curve or AUC) may be normal or most often elevated in absolute terms, but should be regarded as reduced considering the elevated glycemic values. Moreover, the insulin response during OGTT may show a sluggish initial response and elevated values in the later stages, the latter perhaps resulting 26Belfiore/Iannello Pharmacological Stimulation of Insulin Secretion Insulin Secretion as Modified by Sulfonylureas The main drugs able to stimulate insulin secretion are the sulfonylureas. These compounds have been used in the management of type 2 diabetes since 1955 and, when properly utilized, are easy to use and appear to be effective and safe. It is estimated that 30–40% of diabetic patients are taking oral sulfonylureas. Indications and contraindications for sulfonylureas are shown in tables 1 and 2, respectively. Table 1. Patients candidate for sulfonylurea treatment Most patients with type 2 diabetes, not well controlled with dietary restriction and exercise Children and adults with the MODY (maturity-onset diabetes of youth) type of diabetes Obese-diabetic patients with marked insulin resistance Lean type 2 diabetic patients with preserved insulin secretory capacity Table 2. Contraindications to sulfonylurea treatment Patients with type 1 diabetes Patients with pancreatic diabetes Patients with an acute illness or stress or undergoing surgery Patients with hepatic or liver diseases Patients predisposed to hypoglycemia: Underweight or malnourished Elderly Diabetic pregnancy: Potential teratogenicity Perinatal mortality Severe neonatal hypoglycemia Diabetic female patients during lactation Patients with a history of severe adverse reactions to sulfonylureas Different Sulfonylureas The first oral hypoglycemic drug was synthesized in 1926 by altering the guanidine molecule. The sulfonylureas used today are derived from this native molecule. The ‘first-generation’ sulfonylureas, which were developed initially, are effective in large doses, while the ‘second-generation’ drugs, developed more recently, are effective in smaller doses. Some sulfonylureas, such as tolbutamide, 29Insulin Secretion and Its Pharmacological Stimulation Table 3. Main characteristics of sulfonylureas Compound Dose, mg/day Doses q.d. Duration of Metabolism/ hypoglycemic excretion effect, h First generation Acetohexamide 250–1,500 1–2 12–18 Liver/kidney Tolbutamide 500–3,000 2–3 6–12 Liver Chlorpropamide 100–150 1 60 Kidney Tolazamide 100–1,000 1–2 12–14 Liver Second generation Glibenclamide (or glyburide) 1.25–20 1–2 16–24 Liver/kidney Glyburide, micronized 0.75–12 1–2 12–24 Liver/kidney Glipizide 2.5–40 1–2 12–24 Liver/kidney Gliclazide 80–320 1–3 10–20 Liver/kidney Gliquidone 30–120 1–3 6–12 Liver Glimepiride 1–8 1 D24 Kidney Repaglinide1 0.5–16 1–4 4–6 Liver 1 Repaglinide is a nonsulfonylurea hypoglycemic agent of the meglitinide family. have a short duration of action (6 h), others, such as chlorpropamide, have a long duration of action (up to 60 h), several others show an action of inter- mediate duration. Some characteristics of the sulfonylureas which are or have been in clinical use are summarized in table 3. ‘First-Generation’ Sulfonylureas. Tolbutamide has a ‘short’ duration of action (see table 3) and is carboxylated by the liver to a totally inactive derivative. Being metabolized only in the liver, this compound may be useful in nephropathic diabetic patients. Tolazamide has a more potent hypoglycemic activity than tolbutamide and an ‘intermediate’ duration of action (see table 3). It is metabolized only by the liver with the production of some very little active metabolites excreted in the urine (85%). It is safer in the elderly and in nephropathic diabetic patients. Tolazamide also has a diuretic action. Chlorpropamide has a more potent hypoglycemic activity than tolbuta- mide and a ‘very long’ duration of action (see table 3), and therefore it can induce more hypoglycemic episodes than tolbutamide. It is hydroxylated by the liver with production of some active metabolites excreted in the urine (by 80–90%) and, thus, is contraindicated in the elderly and in nephropathic diabetic patients. Several side or toxic effects may occur with chlorpropamide, 30Belfiore/Iannello such as alcohol-induced flushing, occasional hypersensitivity reactions as well as water retention and hyponatremia (due to sensitization of renal tubules to antidiuretic hormone). Acetohexamide has a more potent hypoglycemic activity than tolbutamide and an intermediate duration of action. It is reduced by the liver to 1-hydroxy- hexamide which is a potent hypoglycemic drug, excreted by 60% in the urine. Thus, it is contraindicated in the elderly and in nephropathic diabetic patients. Acetohexamide also has diuretic and uricosuric actions. ‘Second-Generation’ Sulfonylureas. Glyburide or glibenclamide has been used since 1969. It has a 50–100 times more potent hypoglycemic activity than the ‘first-generation’ drugs and has a relatively long duration of action. It is metabolized by the liver to several both inactive and mildly active metabolites, excreted partially in the urine (50%) and partially in the bile (50%). It may induce severe hypoglycemic episodes and is contraindicated in the elderly and in nephropathic diabetic patients. Glyburide absorption is not affected by food but it takes 30–60 min to achieve adequate plasma levels, so that this drug should be taken before the morning meal. Glipizide has been used since 1973, has a 50–100 times more potent hypoglycemic activity than the ‘first-generation’ drugs (comparable to that of glyburide) and has an ‘intermediate’ duration of action (see table 3). It is metabolized by the liver to several inactive metabolites, excreted in the urine (by 68%) and in the feces (by 10%). It may induce severe hypoglycemic episodes (similarly to glyburide) and is contraindicated in the elderly and in nephro- pathic diabetic patients. The absorption of glipizide is delayed by about 30 min when it is ingested with a meal, so that it is recommended to take the drug 30 min before meals. Glipizide has a greater effect than glyburide in raising postprandial plasma insulin level and lowering postprandial plasma glucose level while glyburide has a better effect than glipizide in raising fasting insuline- mia and reducing fasting glycemia (probably, reducing fasting hepatic glucose production). For this metabolic difference, a ‘combined’ administration of the two sulfonylureas was suggested. Gliclazide has a potent hypoglycemic activity (comparable to that of glyburide and glipizide) and has an ‘intermediate’ duration of action. It is metabolized by the liver to several probably inactive metabolites, excreted in the urine (by 60–70%). It has been suggested that gliclazide exerts antiplatelet aggregating activity, with a potential preventing effect on diabetic microangi- opathy, although this effect has not been confirmed. Gliquidone has a short duration of action (the mean half-life was approxi- mately 1.2 h and the mean terminal half-life was 8 h), is metabolized in the liver to totally inactive or minimally active derivatives, and is excreted in the intestine (by about 100%). For these reasons, gliquidone is safer in the elderly 31Insulin Secretion and Its Pharmacological Stimulation Table 4. Extrapancreatic effects of sulfonylureas Hormonal effects Potentiation of insulin action on skeletal muscle and adipose tissue glucose transport Potentiation of insulin action on hepatic glucose production (activation of glycogen synthase and glycogen synthesis) Decrease of hepatic insulin extraction Decrease of insulin degradation (inhibition of insulinase activity) Stimulating effect on gastrointestinal hormone release Direct metabolic effects Insulin receptors (partial restoration of their number in plasma membrane in type 2 obese- diabetic patients) Liver (increase in fructose 2,6-bisphosphate; increase in glycolysis; decrease in gluconeogen- esis; decrease in long-chain fatty acid oxidation) Skeletal muscle (increase of glucose and amino acid transport; increase of fructose 2,6- bisphosphate) Myocardial tissue (increase of contractility; increase of oxygen consumption; increase of glycogenolysis; decrease of Ca2+-ATPase; increase of glucose transport and glycolysis; increase of phosphofructokinase activity and pyruvate oxidation) Adipose tissue (increase in glycogen synthase; inhibition of lipolysis, increase in glucose transport) Platelet arachidonic acid metabolism (inhibition of cycloxygenase and 12-lipoxygenase path- ways) Other Effects. Sulfonylurea treatment does not appear to stimulate proin- sulin biosynthesis. On the other hand, studies performed with in vivo and in vitro animal perfused pancreases, or with isolated perifused islets and islet- cell cultures, reported an acute and chronic sulfonylurea-induced inhibition of the biosynthesis of proinsulin (through unknown mechanisms). Sulfonylureas, acutely or chronically, do not alter glucagon secretion both in normal subjects and diabetic patients. Sulfonylureas appear to stimulate pancreatic d-cell soma- tostatin release (with unclear physiological effect). Extrapancreatic Effects of Sulfonylureas. Diverse in vitro and in vivo extrapancreatic effects of sulfonylureas have been reported over the last 30 years (most of which, however, were obtained with drug concentrations larger than those achieved in therapeutic use) (table 4). These effects of sulfonylureas are due to direct actions on liver and/or muscle and, occurring in the absence of changes in insulin binding, are probably mediated by postreceptor events. As a whole, the extrapancreatic effects of sulfonylureas are of minor clinical significance. A possible exception is glimepiride, which may exert more signifi- cant extrapancreatic actions, including activation (through dephosphorylation) of GLUT-4. 34Belfiore/Iannello Table 5. Sulfonylurea side or toxic effects Hematologic reactions (agranulocytosis, bone marrow or red cell aplasia, hemolytic anemia) Skin reactions (rash, pruritus, erythema, purpura, photosensitivity) Hypersensitivity reaction (rush, fever, arthralgia, angiitis, jaundice, etc.) Alcohol-induced flushing (most frequently associated with chlorpropamide treatment) Gastrointestinal complaints (nausea, vomiting, jaundice or hepatitis or cholestasis) Antithyroid activity Diuretic effect or antidiuresis with hyponatremia Cataract formation (reported in some dogs treated with high doses of glimepiride) Teratogenicity Side or Adverse Effects of Sulfonylureas. The most important adverse effect of sulfonylureas is hypoglycemia which, although occurring less often than with insulin, when it occurs it tends to be more severe, prolonged and sometimes fatal. The incidence of sulfonylurea-induced hypoglycemia is 0.19–4.2/1,000 treatment years (compared to 100/1,000 patients/year for insulin-induced hy- poglycemia) and is most frequent in patients taking long-acting drugs (such as glyburide and chlorpropramide) which, for this reason, should be avoided in patients with predisposing conditions (the best treatment of hypoglycemia is prevention). The case fatality rate of hypoglycemia induced by sulfonylureas is 4.3% (see also chapter VIII on Clinical Emergencies in Diabetes. 2: Hypogly- cemia). It is noteworthy that sulfonylureas predispose to hypoglycemia during and after exercise. In this regard, it has been claimed that glimepiride maintains a more physiological regulation of insulin secretion during physical exercise, with less risk of hypoglycemia. Other sulfonylurea side effects or toxic reactions occur at low rate (1.5% for glyburide) (table 5) and appear within the first 2 months of treatment. The chlorpropamide alcohol flushing (CPAF), occurring in 30–40% of type 2 and 10% of type 1 diabetic patients, is linked to a genetic predisposition to diabetes development (autosomic trait) and can be considered a good genetic marker of type 2 diabetes mellitus. Other Drugs Modifying Insulin Secretion Repaglinide is a nonsulfonylurea hypoglycemic agent of the meglitinide family, a new class of drugs with insulin secretory capacity which exert a rapid- and also short-acting effect, thus entailing reduced risk of long-lasting, and hence dangerous, hypoglycemia. Repaglinide appears to bind to receptor 35Insulin Secretion and Its Pharmacological Stimulation sites different from those of sulfonylureas (two binding sites have been identi- fied). Repaglinide lowered fasting and postprandial blood glucose levels in animals, healthy volunteers and patients with type 2 diabetes mellitus. Repagli- nide is rapidly absorbed and eliminated, which may allow a relatively fast onset and offset of action. Excretion occurs almost entirely by nonrenal mecha- nisms. In comparative clinical trials in patients with type 2 diabetes mellitus, repaglinide 0.5–4 mg twice or 3 times daily before meals provided similar glycemic control to glibenclamide (glyburide) 2.5–15 mg/day. Addition of repaglinide to existing metformin therapy resulted in improved glycemic con- trol. In contrast with glibenclamide, use of repaglinide allowed patients to miss a meal without apparently increasing the risk of hypoglycemia. GLP-1 has insulinotropic action, which may explain the increased insulin response after oral compared to intravenous glucose administration, and exerts several other functions such as reduction of glucagon concentration, reduction of gastric emptying, stimulation of proinsulin biosynthesis and reduction of food intake (upon intracerebroventricular administration in animals). On these grounds, GLP-1 seems to offer an interesting perspective in treatment of diabetic patients. The observations that GLP-1 induces both secretion and production of insulin, and that its activities are mainly glucose-dependent, led to the suggestion that GLP-1 may present a unique advantage over sulfonylurea drugs in the treatment of type 2 diabetes. This peptide is able to lower and perhaps normalize fasting hyperglycemia and to reduce postprandial glycemic increments (especially in type 2 diabetic patients) but its usefulness is not completely established. Due to rapid proteolytic cleavage, the half-life of GLP-1 is too short for therapeutical use with subcutaneous injections. GLP-1 analogues with different pharmacokinetic properties (or some preparations that could be orally administered) are in development. Given the large amount of GLP-1 present in L-cells, it appears worthwhile to look for some agents that could ‘mobilize’ this endogenous pool of the ‘antidiabetogenic’ gut hormone GLP-1. Interference with sucrose digestion using a-glucosidase inhibition moves nutri- ents into distal parts of the gastrointestinal tract and, thereby, prolongs and augments GLP-1 release. Antiarrhythmic agents with Vaughan Williams class Ia action have been found to induce a sporadic hypoglycemia. Recent investigation has revealed that these drugs induce insulin secretion from pancreatic b-cells by inhibiting ATP-sensitive K+ (K-ATP) channels in a manner similar to sulfonylurea drugs. It is possible that in the future, pharmacological compounds will be found that may act on GK and improve b-cell insulin secretion. 36Belfiore/Iannello Fig. 1. Schematic representation of dose-response curves of insulin action in the normal state and in conditions of impaired insulin action. For explanation, see the text. When the receptor number is decreased, the number of insulin molecules that bind to the receptors at a given insulin level will be reduced, and therefore the insulin effects will be diminished, i.e. there is insulin resistance. However, by increasing the insulin level, the number of insulin molecules that bind to the receptors can be increased toward the normal and therefore the insulin effects can be restored; moreover, by increasing further the insulin level, the maximum effect can be reached. This condition is called decreased insulin sensitivity. By plotting the insulin concentrations (on the abscissa) against the insulin effect (on the ordinate), the insulin dose-response curve is obtained. This curve, in the case of insulin resistance due to reduced receptor number, will be shifted to the right, as the maximum effect is reached at very high insulin levels. On the other hand, when the insulin resistance is due to defects in postreceptor steps of insulin action (see below), the dose-response curve is flattened and the maximum insulin effect is not reached even at very high insulin concentrations. When the two conditions coexist, the insulin dose- response curve will be shifted to the right and flattened (fig. 1). Concerning the fate of the insulin-receptor complexes, several data suggest that they are internalized and delivered to endosomes, the acidic pH of which induces the dissociation of insulin molecules from insulin receptors and their sorting in different directions: insulin molecules are targeted to late endosomes 39Insulin Resistance and Its Relevance to Treatment and lysosomes where they are degraded whereas receptors are recycled back to the cell surface in order to be reused. To understand the function of insulin receptors, it should be recalled that protein kinases that directly phosphorylate proteins are divided into two major classes: those that phosphorylate tyrosine (tyrosine-specific protein kinases) and those that phosphorylate serine and threonine (the serine/threonine-spe- cific protein kinases). The receptor b-subunit can be phosphorylated on serine, threonine and tyrosine residues and possesses intrinsic protein-tyrosine kinase activity. Insulin stimulates this activity (i.e. the insulin receptor is itself an insulin-sensitive enzyme) which is responsible for both autophosphorylation of the receptor itself and phosphorylation of tyrosine residues of various cellular substrates, including the insulin receptor substrates (IRS-1 and IRS-2). The latter, through a mechanism not yet fully understood, trigger a sequence of events which include phosphorylation/dephosphorylation of several cyto- plasmatic proteins which, in turn, will induce the spectrum of insulin effects (fig. 2). Two insulin receptor isoforms have been identified, the A and the B form, which, however, revealed no difference in their tyrosine kinase activity in vivo. Protein-tyrosine phosphatases (PTPases) play an essential role in the regu- lation of reversible tyrosine phosphorylation of cellular proteins that mediate insulin action. In particular, some data suggest a possible role of the transmem- brane PTPase in insulin receptor signal transduction. Recent studies suggest that the insulin receptor tyrosine kinase inhibitor, the membrane glycoprotein PC-1, may modulate insulin activity (and may play a role in insulin resistance – see the second part of this chapter). Metabolic Effects of Insulin (Postreceptor Effects) The mechanisms of postreceptor insulin effects can be distinguished into: (a) translocation (and activation) of glucose transporters (the GLUT-4 iso- form) from the intracellular pool to the cell membrane; (b) activation/inhibition of several enzymes of intermediary metabolism through either changes in concentrations of ions or regulatory compounds which bind to the enzyme at sites distinct from the substrate-binding site (allosteric effectors), or covalent modifications of the enzyme molecules often consisting of phosphorylation/ dephosphorylation processes; (c) induction/repression mechanisms leading to changes in enzyme concentration through regulation of the synthesis of the enzyme proteins. Translocation and activation/inhibition processes are short- term mechanisms (occurring within seconds or minutes), the induction/repres- sion processes are long-term mechanisms (hours). 40Belfiore/Iannello Stimulation of glucose transport across the cell membrane is one of the main effects of insulin in muscle and adipose tissue, and is the result of the translocation of glucose transporter (the GLUT-4 isoform)-containing vesicles from an intracellular storage pool to the surface membrane. This event is mediated through IRSs, which in turn activate PI-3-kinase isoforms (fig. 2). Translocation and activation of GLUT-4 is favored by its dephos- phorylation. In addition to glucose transport, insulin also stimulates the transport across the cell membrane of amino acids and ions, mainly potassium and phosphate. Insulin regulates several key metabolic steps (fig. 1). In doing so, insulin is opposed by the four counterregulatory hormones (the rapid-acting glucagon and catecholamines, and the slow-acting growth hormone and cortisol). Insulin affects the pathways of glucose utilization as well as the synthesis and degrada- tion of macromolecules (glycogen, triglycerides and proteins) by regulating the activity of ‘key enzymes’. Indeed, along each metabolic pathway, there is one or more key step(s) catalyzed by key enzymes. These are enzymes which, because of their low activity and sensitivity to regulatory factors (including hormones), regulate the overall rate of the pathway to which they belong. In particular, insulin (or, better, its prevalence over the counterregulatory hormones) exerts the following effects (fig. 1): (a) favors glucose utilization by activating the three key glycolytic kinases, namely hexokinase (and GK in the liver), phosphofructokinase and pyruvate kinase; in the liver, this is associated with repression of the opposing key gluconeogenic enzymes: glucose-6-phosphatase, fructose bisphosphatase and phosphoenolpyruvate carboxykinase plus pyruvate carboxylase; (b) stimulates glucose oxidation, by activating the key enzyme pyruvate dehydrogenase in the mitochondria; (c) lowers FFA level by inhibiting lipolysis in the adipose tissue and reduces ketogenesis from FFA in the liver (see chapter VII on ketoacidosis for further explanation); (d) favors glycogen synthesis and depresses glycogenolysis by activating the enzyme glycogen synthase while inhibiting glycogen phosphorylase; (e) enhances triglyceride synthesis and refrains triglyceride hydrolysis (lipolysis) by inhibiting the hormone sensitive lipase; (f ) finally, insulin stimulates protein synthesis and opposes protein deg- radation (or proteolysis). The main insulin actions are summarized in table 1. Thus, the overall action of insulin is (1) to increase glucose utilization in muscle, liver and adipose tissue while depressing glucose production in the liver, which results in blood glucose lowering; (2) to lower FFA level by refraining lipolysis, and (3) to prevent ketone formation in the liver by opposing ketogenesis. 41Insulin Resistance and Its Relevance to Treatment Insulin Resistance in Type 2 Diabetes and Obesity In type 2 diabetic patients, insulin resistance is due to impaired insulin action either at receptor and postreceptor level, and may result from two etiological components, the genetic background and some acquired factors, of which overweight and obesity are certainly the most important ones. Insulin Receptor Defects A common cause contributing to decreased insulin action consists of reduction in the insulin receptor number which, however, most often is second- ary to insulin resistance and the associated hyperinsulinemia, through the ‘downregulation’ mechanism. In type 2 diabetes, an incomplete activation of the insulin receptor tyrosine kinase appears to contribute to the pathogenesis of the signalling defect. Available data suggest that the impaired tyrosine kinase function of the insulin receptor is not due to an inherited defect but rather is caused by a modulation of insulin receptor function. The B isoform is increased in the skeletal muscle in type 2 diabetes, which may not have significant functional significance. In this context, it is worthwhile noting that in obese subjects, increased PTPase activity has been found in the adipose tissue that can dephosphorylate and inactivate the insulin receptor kinase. The membrane glycoprotein PC-1 (PC-1) has been proposed to be an ecto-protein kinase capable of phosphorylating itself as well as exogenous proteins, and would act as an inhibitor of the tyrosine kinase activity of the insulin receptor. PC-1 was found to be increased in tissues (muscle and fibroblasts) of insulin-resistant subjects. Moreover, in transfected cell lines that overexpress PC-1 there is a reduction in the insulin-stimulated insulin receptor tyrosine phosphorylation. These and other data raise the possibility that PC-1 has a role in the insulin resistance of noninsulin-dependent diabetes mellitus as well as of obesity. In obese patients, skeletal muscle shows reduction in the phosphorylation of insulin receptor and IRS-1 and in PI-3-kinase activation. The scarce ex- pression of these proteins would contribute to determine muscular insulin resistance. Hyperglycemia might directly inhibit insulin-receptor tyrosine kinase ac- tivity and the receptor function. This appears to be mediated by activation of certain protein kinase C isoforms which form stable complexes with the insulin receptor and modulate the tyrosine kinase activity of the insulin recep- tor through serine phosphorylation of the receptor b-subunit. 44Belfiore/Iannello Postreceptor, Metabolic Mechanisms Key Metabolic Steps Resistant to Insulin. Concerning glucose metabolism, by means of complex procedures including the glucose clamp technique, la- belled compound infusion and indirect calorimetry, it has been shown that in patients with type 2 diabetes there is an impairment of insulin-mediated glucose utilization by peripheral tissue (muscle) as well as a reduced insulin suppression of hepatic glucose production. Thus, reduction in insulin sensitivity occurs for both the peripheral glucose utilization and the hepatic glucose production. Reduction in the receptor number may play a role, although it is more probable that this is a secondary phenomenon due to the downregulation of receptors by the hyperinsulinemia which accompanies insulin resistance. A defect in intracellular dissociation of the insulin-receptor complex might contribute by altering the receptor recycling and insulin processing. However, especially in the most severe cases, reduction in insulin responsiveness of peripheral glucose utilization does occur, suggesting postbinding defects in insulin action. Impairment of insulin-mediated glucose utilization by peripheral tissue (muscle) is due mainly to reduction in nonoxidative glucose utilization (glyco- gen synthesis) and to a minor extent to reduced glucose oxidation. Concerning nonoxidative glucose utilization (glycogen synthesis), the defect has been ten- tatively located at the level of the key enzyme glycogen synthase, or the glyco- gen-synthase-activating enzyme, protein phosphatase-1. Other data suggest the implication of earlier steps of glucose utilization, such as glucose transport and/or glucose phosphorylation to glucose-6-P (effected by the enzyme hexoki- nase), which may secondarily impair glycogen synthase activation. On the other hand, the defect in glucose oxidation has been located at the level of the pyruvate dehydrogenase reaction. Concerning lipid metabolism, resistance of lipolysis to the antilipolytic ac- tion of insulin often occurs in type 2 diabetic patients, especially when overweight or obesity is present, which results in elevated FFA plasma levels. However, it should be noted that, when obesity is present, elevation of plasma FFA may also be due to the increased fat mass, even in the presence of normal lipolysis. Insulin suppresses VLDL production in insulin-sensitive humans partly by suppressing plasma FFA levels and partly by a non-FFA-mediated, direct he- patic mechanism (inhibition of ApoB synthesis). Insulin-resistant hyperinsuli- nemic obese individuals are resistant to this suppressive effect of insulin on VLDL-ApoB production. Resistance to the normal suppressive effect of insulin, in addition to other metabolic abnormalities associated with insulin resistance, may contribute to postprandial and postabsorptive hypertriglyceridemia. When obesity is present, insulin resistance of fat storage may also be present, which may be an adaptation limiting further fat deposition, but is maladaptive in terms of risk factors for atherosclerosis. 45Insulin Resistance and Its Relevance to Treatment Genetic Factors (Changes in Predisposed Individuals). The genetic compo- nent of the insulin resistance is suggested by the observation that in first- degree relatives of type 2 patients, the insulin-stimulated glucose metabolism was reduced, which was accounted for by a defect in nonoxidative glucose utilization (glucose storage as glycogen), whereas glucose oxidation rate ap- peared normal. Consistently, impaired activation of glycogen synthase by insulin has been reported in these individuals. The glycogen-synthase-activat- ing enzyme, protein phosphatase-1, also has an abnormally low level of activity in these subjects. Moreover, in insulin-resistant offspring of parents with nonin- sulin-dependent diabetes mellitus, after muscle glycogen-depleting exercise, there is a severely diminished rate of muscle glycogen synthesis during the recovery period (2–5 h), which is known to be insulin-dependent. These data, however, do not necessarily mean that the defect is located at the glycogen synthase level, since defects in the earlier step of glucose metabolism, such as transport and/or phosphorylation, may also impair glycogen synthase activa- tion. Acquired Factors: The Reciprocal Negative Influences of FFA and Glucose Metabolism. Among the acquired factors favoring insulin resistance, obesity, which results from absolute or relative hyperphagia and/or hypoactivity, is certainly the most important one. Obesity is associated with elevated FFA levels, as well as with enhanced availability of glucose (hyperphagia) and insulin. In these conditions, there is a competition between FFA and glucose as energetic fuels. Indeed, reciprocal negative influences between FFA and glucose metabolism occurs through the mechanisms outlined below (fig. 3). In fact, the utilization of FFA (tendencially enhanced in obesity, because of the trend to high plasma FFA levels) leads to the formation of long-chain CoA or acyl-CoA (LC-CoA) in the cytosol, followed by the entry of LC-CoA into the mitochondria through the action of the enzyme carnitine palmitoyl transferase-1 (CPT-1) and by the b-oxidation of LC-CoA to acetyl-CoA. Both these compounds inhibit glucose utilization, thus inducing insulin resistance. In fact: (a) acetyl-CoA inhibits the key enzyme of glucose utilization, pyruvate dehydrogenase; the resulting inhibition of oxidative glucose utilization will maintain saturated the glycogen stores, thus refraining further glucose conver- sion to glycogen, i.e. the nonoxidative glucose utilization; (b) LC-CoA (directly or through the generation of regulatory metabolites) exerts complex metabolic effects, including inhibition of nonoxidative glucose metabolism (fig. 3). On the other hand, the metabolism of glucose (favored in obesity by dietary carbohydrates and hyperinsulinism) leads to the formation of malonyl-CoA (fig. 3), a regulatory compound which inhibits CPT-1, thus reducing the intra- mitochondrial transport of LC-CoA and therefore its b-oxidation. This favors the accumulation of LC-CoA in the cytosol. 46Belfiore/Iannello receptor (IR) and of its major insulin receptor substrates, IRS-1 and IRS-2, which may be a molecular mechanism for uncoupling insulin signaling, as enhanced Ser/Thr phosphorylation of IRS-1 and IRS-2 impairs their inter- action with the juxtamembrane region of IR. Thus, the TNFa produced by adipocytes may function as a local ‘adipostat’ to limit fat accumulation. Increased production of TNFa by fat cells stimulates downregulation of the insulin-sensitive glucose transporter, GLUT-4, in adipocytes. TNFa is overex- pressed in the adipose tissue of obese rodents and humans, and is associated with insulin resistance. The exact role of TNFa, however, remains to be estab- lished. Leptin. Leptin is the product of OB gene. This 16-kDa protein is produced by mature adipocytes and is secreted in the plasma. Its plasma levels are strongly correlated with adipose mass in rodents as well as in humans. Leptin inhibits food intake, reduces body weight and stimulates energy expenditure. Leptin binds to a long-form of leptin receptor in the hypothalamus, thus stimulating the release of GLP-1 and decreasing the production of neuropep- tide Y, a neuromediator (stimulator) of food intake. Recent studies have shown that leptin inhibits insulin secretion and has anti-insulin effects on liver and adipose tissue. If these effects are confirmed, leptin could play a role similar to that of TNFa and could participate in the insulin resistance of obesity and type 2 diabetes. Serum leptin is increased in insulin-resistant offspring of type 2 diabetic patients. Other Factors Contributing to Insulin Resistance Decreased blood flow and capillary density has been proposed as mecha- nisms contributing to insulin resistance both in type 2 diabetes and in obese insulin-resistant Pima Indians. It has been suggested that insulin action may be modulated by blood flow. Insulin resistance in moderately obese women was associated with an abnormal vascular reactivity to stress, entailing exaggerated blood pressure response; an enhanced vasoconstriction to stress may mediate this response. Insulin-induced attenuation of noradrenaline-mediated vasoconstriction is impaired in the obese rats. This defect in insulin action could reside in the endothelial generation of nitric oxide, as endothelial function is also abnormal. A distinct capillary endothelial dysfunction may be involved in the insulin resistance syndrome. However, the capillary wall crossing is rate-limiting for muscle glucose uptake (and lactate release) in control subjects but not in postabsorptive hyperglycemic insulin-resistant subjects. On the other hand, in a prospective (15 years) study it was found that capillary density was increased rather than decreased in subjects with impaired 49Insulin Resistance and Its Relevance to Treatment glucose tolerance who later developed diabetes, a fact that might be regarded as a compensatory mechanism for intracellular defects in glucose metabolism. In healthy young men, there is a negative relationship between directly measured whole-blood viscosity and insulin sensitivity (clamp technique) as a part of the insulin resistance syndrome, which supports the hypothesis that insulin resistance has a hemodynamic component. Insulin-resistant first-degree relatives of type 2 diabetic patients were shown to have an increased number of the glycolytic, fast-twitch (white), type IIb muscle fibers compared to the oxidative, slow-twitch (red), type I and IIa fibers (which are those normally responsive to insulin). Whether this finding reflects a reduced physical activity level and fitness in the relatives or is of primary genetic origin remains to be determined. Insulin Resistance in Type 1 Diabetes Although type 1 diabetes is due to severe insulin deficiency, it should be considered that chronic lack of insulin action may produce insulin resistance, through several mechanisms. Asalready pointedout, insulinexerts bothshort-termand long-termeffects, the latter consisting of induction or repression of the synthesis of key enzymes. Therefore, in the tissues of the insulin-deficient subject there will be a decreased content of the enzymes ‘induced’ by insulin (example: hepatic GK) and accumu- lation of the enzymes repressed by insulin (example: gluconeogenic enzymes). Upon insulin administration, some degree of ‘resistance’ (reduced effect) may occur until the enzyme balance is normalized, i.e. until the amount of those enzymes ‘induced’ by insulin is restored and the accumulation of those enzymes ‘repressed’ by insulin decreases to the normal level, which may takes several hours or even 1–2 days. In addition, severe insulin deficiency is always associated with active lipolysis and enhanced release of FFA, which counteract insulin ac- tion through the mechanism of the glucose-FFA cycle, already discussed. When the type 1 diabetic patient becomes severely decompensated and ketoacidosis supervenes, the insulin action may be further disturbed by the interference of the acidosis with insulin binding to its receptor as well as by the reduced response of the intracellular enzymes caused by the hyperosmolality. Rare Genetic Forms of Insulin Resistance These are described in chapter I on Etiological Classification, Pathophysi- ology and Diagnosis. 50Belfiore/Iannello Drugs Ameliorating Insulin Resistance Introduction The recognized major role played by insulin resistance in the pathogenesis of type 2 diabetes is the rational basis for the use of drugs capable of improving insulin sensitivity and consequently enhancing insulin action. The most used of these drugs today is the biguanide compound metformin (dimethylbiguan- ide), but other potentially useful agents, today under clinical investigation, will also be considered, such as the thiazolidinediones (pioglitazone, troglita- zone and rosiglitazone) and still others. Metformin The main biguanides (phenformin and metformin) were first synthesized in 1929 and were shown to be potent antihyperglycemic agents. They were rediscovered in 1957 and were widely used in Europe to treat obese type 2 diabetic patients. Phenformin was withdrawn in many countries because of an association with lactic acidosis, but metformin, which does not bear the same risk when appropriately prescribed, resurfaced in the 1980s and was shown to increase insulin sensitivity. This led to its approbation for use in the USA in 1994. Actions. Unlike sulfonylureas (and the biguanide phenformin), metformin does not bind to plasma proteins and is not metabolized by the liver. Metfor- min has an absolute oral bioavailability of 40–60%, and gastrointestinal absorption is apparently complete within 6 h of ingestion. An inverse relation- ship was observed between the dose ingested and the relative absorption with therapeutic doses ranging from 0.5 to 1.5 g, suggesting the involvement of an active, saturable absorption process. Metformin is eliminated rapidly by the kidney and has a mean plasma elimination half-life after oral administration of between 4.0 and 8.7 h (approx. 6 h). The elimination is prolonged in patients with impairment of renal function and correlates with creatinine clearance. Therapeutic blood levels may be 0.5–1.0 mg/l in the fasting state and 1–2 mg/l after a meal. However, monitoring of blood levels may be useful only to confirm the diagnosis of lactic acidosis. Metformin has no effect in the absence of insulin, because the drug seems to act primarily by enhanc- ing insulin action at postreceptor level. Metformin ameliorates hyperglycemia by improving peripheral sensitivity to insulin and reducing hepatic glucose production (via gluconeogenesis) as well as by limiting gastrointestinal glu- 51Insulin Resistance and Its Relevance to Treatment muscles, adipose tissue and hepatocytes, while normalizing a wide range of metabolic abnormalities associated with insulin resistance. Reported effects include: (a) decrease in plasma triglyceride, FFA and LDL cholesterol levels and increase in plasma HDL cholesterol; (b) increased expression of glucose transporters GLUT-1 and GLUT-4; (c) activation of glycolysis in hepatocytes; (d) antagonism towards some of the effects of TNFa; (e) decrease in blood pressure; (f ) inhibition of vascular smooth muscle cell proliferation and hypertrophy; (g) enhanced endothelium-dependent vasodilation, and (h) anti- oxidant action. Finally, although thiazolidinediones do not stimulate insulin secretion, they improve the secretory response of b-cells to insulin secretagog- ues. Rosiglitazone (a PPARc Agonist). Rosiglitazone, like other thiazolidine- dione compounds, is a PPARc agonist, inasmuch as it potently and specifically stimulates peroxisome proliferator-activated receptors-c (PPARc) and sensi- tizes cells to insulin. Indeed, rosiglitazone is an antidiabetic agent which en- hances sensitivity to insulin in the liver, adipose tissue and muscle, resulting in increased insulin-mediated glucose disposal. This compound, therefore, improves insulin resistance, which is a key underlying metabolic abnormality in most patients with type 2 diabetes. In contrast with troglitazone, rosiglita- zone does not appear to be hepatotoxic, on the basis of clinical and in vitro studies, and does not induce cytochrome P450 3A4 metabolism. However, the drug is contraindicated in patients with history or signs/symptoms of liver diseases and its use requires monitoring of liver function tests. Moreover, rosiglitazone does not interact significantly with nifedipine, oral contraceptives, metformin, digoxin, ranitidine, or acarbose. In clinical trials, rosiglitazone 2–12 mg/day (as single daily dose or two divided daily doses) improved glycemic control in type 2 diabetic patients, as shown by decrease in fasting plasma glucose and glycated hemoglobin (HbA1c). Addition of rosiglitazone 2–8 mg/day to existing sulfonylurea, met- formin or insulin therapy achieved reductions in fasting plasma glucose and HbA1c. Consistent with its mechanism of action, rosiglitazone appears to be associated with a low risk of hypoglycemia (=2% of patients receiving mono- therapy) and did not increase the risk of alcohol-induced hypoglycemia. Other Compounds The long-acting, nonsulfhydryl-containing ACE inhibitor, trandolapril, alone and in combination with the Ca2+-channel blocker, verapamil, can sig- nificantly improve whole-body glucose metabolism by acting on the insulin- 54Belfiore/Iannello sensitive skeletal muscle glucose transport system in obese Zucker rats. Data on the role of TNFa raise the possibility that pharmacological inhibition of this factor may provide a novel therapeutic target to treat patients with type 2 diabetes. Suggested Reading American Diabetes Association: Consensus Development Conference on Insulin Resistance, Nov 5–6, 1997. Diabetes Care 1998;21:310–314. Bell PM, Hadden DR: Metformin. Endocrinol Metab Clin North Am 1997;26:523–537. Scheen AJ: Clinical pharmacokinetics of metformin. Clin Pharmacokinet 1996;30:359–371. Daniel JR, Hagmeyer KO: Metformin and insulin: Is there a role for combination therapy? Ann Pharma- cother 1997;31:474–480. Davidson MB, Peters AL: An overview of metformin in the treatment of type 2 diabetes mellitus. Am J Med 1997;102:99–110. DeFronzo RA, Bonadonna RC, Ferrannini E: Pathogenesis of NIDDM: A balanced overview. Diabetes Care 1992;15:318–368. Melchior WR, Jaber LA: Metformin: An antihyperglycemic agent for treatment of type II diabetes. Ann Pharmacother 1996;30:158–164. UK Prospective Diabetes Study (UKPDS) Group: Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet 1998;352:854–865. F. Belfiore, Institute of Internal Medicine, University of Catania, Ospedale Garibaldi, I–95123 Catania (Italy) Tel. +39 095 330981, Fax +39 095 310899, E-Mail francesco.belfiore@iol.it 55Insulin Resistance and Its Relevance to Treatment Chapter IV Belfiore F, Mogensen CE (eds): New Concepts in Diabetes and Its Treatment. Basel, Karger, 2000, pp 56–71 ............................ Diet and Modification of Nutrient Absorption S. Iannello Institute of Internal Medicine, University of Catania, Ospedale Garibaldi, Catania, Italy Diet Introduction In the treatment of diabetes mellitus, changes in lifestyle play a major role, in addition to treatment with insulin or oral glucose-lowering drugs. For most patients with type 2 diabetes, the changes in lifestyle (concerning diet and exercise) are the cornerstone of treatment whereas the pharmacologic intervention represents a supplementary treatment for those patients who do not respond adequately to lifestyle changes. Dietary caloric restriction ameliorates hyperinsulinemia and hyperglyce- mia in obese type 2 diabetics (and improves other metabolic parameters; see table 1) and reduces the incidence of type 2 diabetes in subjects at risk or with impaired glucose tolerance (IGT). Glucose tolerance and insulin sensitivity improve when normal body weight is achieved or approached. Indeed, even a 7–10% of weight loss is enough to improve insulin resistance in all obese type 2 diabetics. Nutritional needs are different in type 1 (lean) or type 2 (overweight or obese) diabetic patients. Diet education is crucial and requires the participation of the patient and its family in the planning-diet process and in the implementation of the adequate strategies to promote adherence to dietary intervention. Goals of dietary therapy in diabetes are to reach and maintain ideal body weight (IBW), to maintain fasting and postprandial glycemic levels as close as possible to normal and to achieve optimal blood lipid values, while providing adequate caloric intake as required for the various metabolic needs. 56 Sedentary normal patients need approximately 30 cal/kg IBW/day while active normal patients need approximately 35–40 cal/kg/day. Overweight sed- entary patients need 20–25 cal/kg/day and active obese patients need 30–35 cal/kg/day, while underweight patients need 35 cal/kg/day if sedentary and 40–50 cal/kg/day if active. In elderly sedentary diabetic patients, 20 cal/kg/day are usually required (after 50 years of age approximately 10% less calories for each decade is required). A more accurate assessment of the caloric needs may be achieved by using appropriate formulas to calculate the rest metabolic rate (RMR), such as those of Harris & Benedict which are based on weight, height, age and sex. Since subjects of the same weight but of different height have similar RMR, formulas may be simplified by considering only weight, age and sex. RMR should be increased by 30, 50 or 70% for low, medium or high levels of physical activity. Table 2 shows the caloric requirement according sex and age for selected weights and activity levels, based on similar formulas. In diabetic children the caloric needs depend on the rate of growth and activity pattern. Children 4–6 years old require 90 cal/kg/day and children 7–10 years old require 80 cal/kg/day. It is important to allow an adequate caloric intake in juvenile diabetes. Caloric requirement in children may also be calculated by adding to the baseline value of 1,000 cal/day the amount of 100–125 cal for every year of age up to 12 years. Youngsters should consume 3 meals daily with 2 or 3 snacks (eaten at the same time each day) to minimize glycemic fluctuations and the risk of hypoglycemic episodes. After the caloric content and the composition of the diet are established, the prescription of a diet was in the past made by utilizing the data in the Exchange Lists for Meal Planning published by the American Diabetes Association. A more useful approach might be to use the precalculated diets (of various caloric content) prepared by several diabetes associations or other authoritative sources. How- ever, it is now recognized that the diet should be individualized and prepared by taking into account the eating habits and other lifestyle factors. It is clinically relevant that 7–35% of adolescent females with type 1 diabetes may have an eating disorder, such as anorexia or bulima nervosa. Dietary Components Dietary Carbohydrate Carbohydrates are the most important source of energy and provide about 4 cal/g. The carbohydrate intake of diabetic patients should be equal to that of nondiabetic subjects. A dietary carbohydrate content of about 50–60% of total energy intake seems adequate in diabetic patients. 59Diet and Modification of Nutrient Absorption Table 2. Caloric needs according to age, sex, weight1 and physical activity Sex and Weight Physical activity age group kg rest rest low low medium medium high high kcal/kg kcal/day kcal/kg kcal/day kcal/kg kcal/day kcal/kg kcal/day Men 18–30 years old 68 25 1,723 33 2,240 38 2,585 43 2,929 72 25 1,784 32 2,319 37 2,675 42 3,032 76 24 1,844 32 2,397 36 2,766 41 3,135 80 24 1,905 31 2,476 36 2,857 40 3,238 Average 74 24.5 1,814 31.9 2,358 36.8 2,721 41.7 3,084 31–60 years old 68 25 1,667 32 2,167 37 2,500 42 2,833 72 24 1,713 31 2,227 36 2,570 40 2,912 76 23 1,760 30 2,288 35 2,639 39 2,991 80 23 1,806 29 2,348 34 2,709 38 3,070 Average 74 23.5 1,736 30.6 2,257 35.3 2,605 40.0 2,952 Women 18–30 years old 56 24 1,323 31 1,720 35 1,985 40 2,249 60 23 1,383 30 1,798 35 2,074 39 2,351 64 23 1,442 29 1,875 34 2,164 38 2,452 68 22 1,502 29 1,953 33 2,253 38 2,553 Average 62 22.8 1,413 29.7 1,836 34.2 2,119 38.8 2,401 31–60 years old 56 23 1,309 30 1,701 35 1,963 40 2,225 60 22 1,342 29 1,744 34 2,012 38 2,281 64 21 1,374 28 1,787 32 2,062 37 2,336 68 21 1,407 27 1,829 31 2,111 35 2,392 Average 62 22.0 1,358 28.6 1,765 33.0 2,037 37.4 2,309 1 Caloric needs at rest (RMR) per day were calculated according to the following formulas (as reported by G. Bray): for 18- to 30-year-old men: (0.0630¶kg weight+2.8957)¶240; for 31- to 60-year-old men: (0.0484¶kg weight+3.6534)¶240; for 18- to 30-year-old women: (0.0621¶kg weight+2.0357)¶240; for 31- to 60-year-old women: (0.0342¶kg weight+3.5377)¶240. RMR was then multiplied by 1.3, 1.5 or 1.7 for low, medium or high physical activity, respectively. Carbohydrates are available as complex or simple sugars. In diabetic patients, complex carbohydrates or polysaccharides should be preferred. Com- plex carbohydrates include: starches (present in large amounts in rice, cereals, potatoes, pulses and vegetable roots), dextrins (derived from hydrolyzed starch), glycogen (contained in liver and muscle), cellulose or pectins (indigest- 60Iannello Table 3. Glycemic index of some foods Bread 100% Beans 65% Rice 83% Grapes 62% Potatoes 81% Apples 53% Bananas 79% Milk 49% Spaghetti 66% Pears 47% Oranges 66% Lentils 43% ible complex carbohydrates contained in plant foods). In diabetics, simple carbohydrates should be restricted. They include monosaccharides (glucose present in oranges and carrots, fructose present in honey and ripe fruits, and galactose derived from hydrolyzed lactose) and disaccharides (sucrose present in beetroot and sugar cane, lactose present in milk, and maltose derived from hydrolyzed starch). The formerly claimed diabetogenic effect of sucrose overconsumption has not been confirmed by epidemiological or experimental studies. However, in diabetic patients, sucrose-rich foods cause a rapid rise in glycemic values, which can be prevented by consuming these foods as part of a mixed meal. The recommended disaccharide (sucrose plus other glucose- containing disaccharides) consumption by diabetic people should not exceed 5–10% of the total caloric intake. Sucrose addition as sweetener should not exceed 20 g/day. Fructose is a natural monosaccharide, used as a sweetener. It is converted to glucose (and stored as glycogen) or triglyceride in liver. In diabetics with insulin deficiency and impaired hepatic glycogen synthesis, fructose-derived glucose contributes to the hyperglycemia. Thus, the safety of fructose use in diabetes is a debated topic. Starches are hydrolyzed to dextrins, then to maltose and finally to glucose (through the effect of gastric acid and intestinal enzymes). They are useful in the diabetic diet because they are slowly digested and absorbed, inducing lower increments of the glycemic and insulinemic values than equivalent amounts of glucose or simple sugars. It is well established that equimolar amounts of carbohydrate in different foods induce different glycemic postprandial excursions. Jenkins et al. [1981] have elaborated a ‘glycemic index’, representing the incremental area under 2 h glycemic curve of food divided by the corresponding area under 2 h glycemic curve after ingestion of a portion of white bread containing equivalent amounts of carbohydrates, multiplied by 100 (table 3). Reference can also be made to the glycemic response after glucose ingestion, in which instance the glycemic index for glucose is 100. Foods containing simple sugars have a high glycemic index, raising glycemia and insulinemia faster and to a greater extent, and therefore are contraindicated in diabetic patients. However, several factors can influence the food glycemic response, including: (a) type of diabetes, age, 61Diet and Modification of Nutrient Absorption secretion, while other amino acids are gluconeogenic and ketogenic. The amount of proteins that should be recommended to diabetic patients depends upon several factors, such as the patient age, the nutritional status (undernutri- tion or malnutrition) and particular situations (growing, pregnancy, lactation, debilitating diseases, nephropathy, uremia, hepatic diseases, etc.). The role of dietary protein in the development and progression of diabetic nephropathy is debated while it is clearly defined that a moderately low protein diet is the best approach for treating renal disease of diabetic patients (see chapter on Diabetic Nephropathy). The recommended amount of proteins in diabetic diet is of 12–20% of total calories. In diabetic subjects a high-protein diet can increase renal blood flow, glomerular filtration rate and intra- glomerular pressure, accelerating glomerulosclerosis to end-stage renal failure (Brenner’s hypothesis). It is useful to substitute, at least in part, vegetable proteins for animal proteins, even if proteins from animal source do not seem to significantly increase kidney workload. In subclinical or incipient stages of diabetic nephropathy, glycemic control and low protein intake (0.8 g/kg IBW/ day) may reduce renal blood flow, restore normal glomerular hemodynamics, decrease proteinuria and delay the progression of nephropathy. In overt diabetic nephropathy with albumin excretion, the recommended protein restriction should be from 0.6 to 0.8 g/kg/day. In cases of protein restriction, essential amino acids should be supplemented. To maintain energy balance, a low protein diet must be high in carbohydrates and fats and may exacerbate hyperglycemia, hypertriglyceridemia or hyperinsulinemia, increasing total and LDL cholesterol and decreasing HDL cholesterol. Moreover, in diabetic pa- tients a low protein dietary content may favor a negative nitrogen balance and muscle wasting. Dietary Fibers In normal subjects and type 2 diabetic patients, several studies demon- strated an improvement of glucose tolerance and a reduction of insulin secre- tion when a diet high in fiber was consumed. In type 1 diabetics, high-fiber diet was found to decrease glycosuria, as well as basal and postprandial glycemic levels. Moreover, high-fıber intake may improve other metabolic parameters, and may also exert a preventive effect on cancer of bowel and diverticular disease (diseases favored by the modern tendency to consume low- fiber, refined foods). Dietary fibers are heterogeneous and consist of several complex polysaccharides resistant to gastrointestinal digestive enzymes (even if certain fibers are metabolized in the colon). Fibers can be water soluble or insoluble and their effects are variable according to the different biochemical- physiological characteristics. Celluloses, hemicelluloses and lignins bind water and cations and are insoluble (wheat products and bran) whereas pectins, 64Iannello Table 5. Foods naturally rich in fibers Legumes Beans, peas, chickpeas, lentils Vegetables Broccoli, artichokes, zucchini, carrots, eggplants, string beans, squash, potatoes, tomatoes, celery, cabbage, onions, beets, fennels, turnips, radishes, asparagus, cucumbers, cauliflower, mushrooms Fruits Apples, blackberries, pears, strawberries, oranges, plums, bananas, grapefruit, pineapples, peaches, cherries, apricots, kiwis, mandarins Cereals Bran (100%), bread (rye), bread (whole-grain wheat), rice, wheat flour (whole grain) gums and mucilages form gels and are soluble (oats, fruits and legumes). The foods naturally rich in fibers are legumes, roots, tubers, whole-grain cereals, fruits and green leafy vegetables (table 5). Usually, the soluble fibers (especially those with high viscosity) exert useful metabolic effects, whereas insoluble fibers contribute to increase fecal bulk, promote movements of intestinal content, being useful in constipation (which may also result from autonomic diabetic neuropathy). The physiological effects of fibers are influenced by osmolality or pH, mixture of fibers and foods, water retention, fermentation by bacteria, etc. Soluble fibers would exert their beneficial effects on carbohydrate and lipid metabolism through several mecha- nisms, which include: (a) satiating effect; (b) delayed gastric emptying time; (c) decreased release of gut hormones, including intestinal insulin secretagogues (as GIP); (d) delayed small intestine transit time and altered colonic emptying time; (e) binding of bile acids, with impaired intestinal absorption of choles- terol; (f ) formation of gels that sequester or hide nutrients (carbohydrates, fats, cholesterol, etc.), providing a physical barrier that separates complex carbohydrates from digestive enzymes, with reduced digestion and absorption in small intestine; (g) increase of fecal bulk with accelerated intestinal transit, which may reduce absorption of nutrients; (h) fermentation by the bacteria in the colon to gases and short-chain fatty acids, which would suppress neoglu- cogenesis, and (i) improved peripheral insulin sensitivity and increased insulin receptor binding. It is interesting that fibers have the best effects when naturally contained in aliments while they have poorer effects when added as pharmaceutical products to dietary foods. Diets useful to improve both fasting and postpran- dial hyperglycemia in diabetic patients have been suggested, which are rich in fibers naturally contained in foods. These diets are very rich in carbohy- drates (up to 70%) and fibers (up to 35 g/day/1,000 kcal, both in soluble 65Diet and Modification of Nutrient Absorption and insoluble forms). In these fiber-rich diets, fibers would mitigate the deleterious effect of the high carbohydrate content on glucose metabolism, and would reduce total or LDL cholesterol and triglycerides (only in dia- betics), while lowering blood pressure and favoring weight loss in obese patients. In hypocaloric diets, the usually recommended fiber supplementation is in the amount of 25 g/1,000 kcal (associated with high water assumption to induce fiber swelling). High-fiber diets can cause (especially in the first 7–10 days) cramping, abdominal discomfort, flatulence and diarrhea. These diets may also impair absorption of minerals and vitamins if used for a long time (in which instance, supplementation of calcium, trace elements and vitamins may be required). They may also increase the risk of bezoar formation, especially when a diet high in fibers is contraindicated (patients with gastrointestinal dysfunction, gastroparesis or altered absorption from pancreatic enzyme deficiency). Large amounts of dietary fibers may not be well tolerated by children, pregnant diabetic women and elderly subjects. Alcohol and Other Nutrients Alcohol provides about 7 cal/g, is not a food but is another source of energy that should be considered in a dietary plan. Interestingly, in women a decreased risk (50%) of developing diabetes with increasing alcohol intake was found and this effect was probably related to lower BMI linked with alcohol consumption. Allowed intake should not exceed 10 g/day. Excessive alcohol intake should be avoided in diabetic patients, because it inhibits glu- coneogenesis and can favor hypoglycemic episodes in subjects treated with insulin or drugs. In hypertriglyceridemic patients, alcohol may exacerbate dyslipidemia and liver steatosis. Diabetic patients may also suffer from associated diseases which require special modified diets. In the presence of congestive heart failure, hypertension and kidney disease, dietary sodium should be restricted. The sodium restriction may range from 500 to 1,000 mg/day (maximum intake =3 g/day), although the use of diuretics may reduce the need for a severe sodium restriction, which makes foods less palatable and may provocate hypotension and fluid or electrolyte disorders. Sweeteners Sweeteners can be distinguished into caloric (or natural) sweeteners and noncaloric (or artificial) sweeteners (table 6). In both type 1 and 2 diabetic patients, the classical sweetener, sucrose, can be allowed in the maximum amount of 20 g/day, especially if associated to a mixed meal, because it does not deteriorate metabolic control. An excessive sucrose intake should be avoided, 66Iannello over IBW and who do not respond to conventional balanced diets, excluding those with recent myocardial infarction, hepatic disease, renal failure or cere- brovascular disease. Monitoring of electrocardiographic changes, urea or creat- inine level and electrolyte disorder is required. Diet and Exercise Exercise is a relevant component in a program of weight loss in diabetic patients. It improves glucose tolerance, lowers glycemia, increases peripheral insulin sensitivity and reduces risk factors for coronary heart disease (amelio- rating hypertension and blood lipid profile). The combination of diet plus exercise is more effective than diet alone or exercise alone in producing long- term weight loss, in maintaining the weight loss over time and in reducing the dose of hypoglycemic drugs. The recommended exercise (walking or stationary bicycle riding) should be of low or moderate intensity but of long duration, and is especially useful in adult or older obese type 2 diabetic subjects. The exercise should be performed at least every 2–3 days for optimum effect (ex- amples: stationary bicycle riding or brisk walking for 30 min/day, or active swimming for 1 h 3 times/week). Because exercise may increase the risk of acute or delayed hypoglycemia, a prospective reduction in insulin dose for regular exercise should be used as well as a supplementary snack of about 40 g of carbohydrates. In decompensated diabetic patients with insulin deficiency, exercise is contraindicated (especially if prolonged, severe or unusual) raising glycemia and ketone levels. Alcohol may exacerbate the risk of hypoglycemia after exercise. Diabetic patients should be encouraged to increase their physical activity gradually, with increments of the activities within their daily lives (walking to work, using stairs rather than elevators, etc.). However, the effects of exercise on the caloric balance (and therefore on weight loss) may be less than expected for several reasons. In fact, from the energy lost during exercise, those calories should be subtracted that the patient would have lost with his usual activity. Moreover, often the exercise stimulates the appetite, leading to enhanced caloric assumption. Finally, in some tense individuals, exercise may induce muscular relaxation, which means decreased isometric muscular work. Conclusion In conclusion, the reduction of the caloric intake in obese people may have a relevant effect on the frequency of type 2 diabetes. On the other hand, a proper nutritional management of obese diabetic patients is the most 69Diet and Modification of Nutrient Absorption important factor of treatment (even if often patients are unable to lose weight or to maintain the reduced weight). A professional consultation with the physician or the dietician is recommended (especially for new patients) at the beginning of diet and then at regular intervals to promote the adherence to dieting and to verify metabolic control of diabetes and weight loss. Dietary adherence is a serious problem in both type 1 and, especially, type 2 diabetic patients. To improve compliance to diet, some strategies were recommended: (a) a meal plan (adequate to the patient’s lifestyle and to the stage of the disease) that involves long-term changes of eating and nutritional habits; periodical reviews of meal plan are needed; (b) the education, which can improve motivation and dietary adherence providing the patient with useful information in an acceptable form to manage effectively nutrition and exercise; (c) a strong feeling between physician and patient (and his relatives for young people), and (d) an interaction afforded by group sessions, in which diabetic patients can exchange experiences and information, providing solutions and behavioral changes through peer example. Modification of Nutrient Absorption Agents capable of modifying the absorption of complex carbohydrates or lipids, such as a-glucosidase inhibitors and Orlistat, will be discussed in Chapter VI (Overview of Diabetes Management). Suggested Reading American Diabetes Association: Nutritional principles for the improvement of diabetes and related com- plications. Diabetes Care 1994;17:490–518. American Diabetes Association: Clinical practice recommendations 1997. Diabetes Care 1997;20(suppl 1): 1–70. Diabetes and Nutrition Study Group of the EASD: Nutritional recommendations for individuals with diabetes mellitus. Diabetes Nutr Metab 1995;8:1–5. Han TS, van Leer EM, Seidell JC, Lean ME: Waist circumference as a screening tool for cardiovascular risk factors: Evaluation of receiver operating characteristics (ROC). Obes Res 1996;4:533–547. International Diabetes Federation (IDF), 1998–1999 European Diabetes Police Group: A Desktop Guide to Type 2 (Non-Insulin-Dependent) Diabetes mellitus. Brussels, IDF, 1999. Jenkins DJ, Wolever TM, Taylor RH, Barker H, Fielden H, Baldwin JM, et al: Glycemic index of foods: A physiological basis for carbohydrate exchange. Am J Clin Nutr 1981;34:362–366. Lemieux S, Prud’homme D, Bouchard C, Tremblay A, Despres JP: A single threshold value of waist girth identifies normal-weight and overweight subjects with excess visceral adipose tissue. Am J Clin Nutr 1996;64:685–693. Perry AC, Miller PC, Allison MD, Jackson ML, Applegate EB: Clinical predictability of the waist-to- hip ratio in assessment of cardiovascular disease risk factors in overweight, premenopausal women. Am J Clin Nutr 1998;68:1022–1027. 70Iannello Rexrode KM, Carey VJ, Hennekens CH, Walters EE, Colditz GA, Stampfer MJ, Willett WC, Manson JE: Abdominal adiposity and coronary heart disease in women. JAMA 1998;280:1843–1848. Vinik A, Wing RR: Nutritional management of the person with diabetes; in Rifkin H, Porte D (eds): Diabetes mellitus. Theory and Practice, ed 4. Amsterdam, Elsevier, 1990, pp 464–496. WHO Expert Committee: Physical Status: The use and interpretation of anthropometry. WHO Tech Rep Ser No 854. Geneva, WHO, 1995. S. Iannello, Institute of Internal Medicine, University of Catania, Ospedale Garibaldi, I–95123 Catania (Italy) Tel. +39 095 330981, Fax +39 095 310899, E-Mail francesco.belfiore@iol.it 71Diet and Modification of Nutrient Absorption Neutral Protamine Hagedorn (NPH) insulin, obtained by adding the protein protamine to insulin and adjusting the pH. (c) The long-acting insulin prepara- tions include the ultralente insulin, obtained by modifying, during the prepara- tion, the pH of a mixture of zinc and insulin to produce larger zinc-insulin crystals (the larger the crystals, the slower the release of the injected insulin) as well as the protamine-zinc insulin obtained by adding also protamine and adjusting the pH. After subcutaneous injection, regular insulin presents a rapid onset of action (0.5–1 h), an early peak of activity (2–4 h) and a duration of action of 4–6 h. Thus, rapid-acting insulin, beginning to act in about 30 min, should be given 20–30 min (perhaps 45 min) before a meal to optimize synchronization of postprandial glycemia and circulating insulin levels. It is effective in blunting elevations in glucose following meals and for rapid adjustments in insulin dosage, but the pharmacokinetics of rapid-acting insulins entails that a definite time interval is observed between insulin injection and eating. A better synchrony between insulin peaks and meal absorption after injection of rapid-acting insulin is observed with human insulin, which acts more rapidly after injection and exerts shorter effects compared to previously used animal insulins. Intermediate-acting NPH insulin presents a delayed onset of action (3–4 h), a delayed peak of activity (8–12 h) and a duration of action of 20–24 h; similar activity is possessed by the lente insulin. The NPH and lente inter- mediate-acting insulins have the same, long time-course of action, which is useful to provide the basal level of insulin through the 24 h when given twice per day. Intermediate-acting human insulin produces earlier peaks, that may cause hypoglycemic events during sleep and fails to maintain an adequate effect for a full 24-hour period. Ultralente (long-acting) insulins present a slow onset of action (6–8 h), a much more delayed peak of activity (14–24 h) and a duration of action of about 32 h. The ultralente human insulin has a shorter duration of action, compared to the animal preparations, and requires also twice-daily injections. Table 2 summarizes the most common insulin preparations and their onset, peak and duration of action. Insulin Analogues Recently, to improve the outcome of insulin therapy and to use human insulin products with more physiological effect, a short-acting monomeric insulin analogue, insulin lispro (Lys[B28], Pro[B29]), was developed which was approved for clinical use and is already commercially available. It has been used extensively in clinical practice. Reversal of the amino acids proline and lysine at position B28 and B29 of human insulin produces an analogue 74Belfiore/Iannello Table 2. Types of insulin Type Preparation Onset of action Peak action Duration of action h h h Rapid-acting Regular 0.5–1 2–4 4–6 Intermediate-acting NPH or lente 3–4 8–12 16–24 Long-acting Ultralente Bovine 4–6 14–24 28–36 Human 4–6 10–20 24–28 There is great variability in the onset, peak and duration of insulin action from patient to patient, as indicated by the time intervals given. Human insulin tends to show somewhat earlier onset and peak and shorter duration of action, compared to nonhuman insulins. This is especially true for long-acting preparations. For this reason, different figures for the time intervals are reported in the table for this insulin preparation. with less tendency to self-association. Conventional insulin preparations are prevailingly in hexameric form, which delays the absorption from subcutane- ous injection sites, requiring the dissociation of hexamers into monomers. The monomeric insulin lispro is rapidly absorbed from subcutaneous tissues (so reducing the postprandial hyperglycemia) and shows a shorter duration of action that should decrease the risk of hypoglycemia between meals and at nighttime. Indeed, it shows early peak (1 h) (which allows a much shorter interval between injection and eating) and a shorter duration of action (3–4 h). The insulin lispro in appropriate dosage may result in a profile of insulin close to the physiological one and is suitable for treating both type 1 and type 2 diabetic patients under intensive insulin therapy. Another rapid- acting insulin analogue is currently under evaluation (B28 Asp). Longer- acting ‘basal’ analogues are also under development such as HOE 901 that is an insulin analogue with a lower peak of activity than NPH and a duration of action very long (about 24 h), especially appropriate for type 1 diabetics. Insulin Concentration The insulin preparation available on the market today is a concentration of 100 U/ml (U-100). In the past (and still today in some countries) a concentra- tion of 40 U/ml (U-40) was in use. However, preparations with more concen- trated insulin also exist (500 U/ml or U-500). 75Insulin Treatment in Type 1 and Type 2 Diabetes Factors Influencing Insulin Concentration or Bioavailability Pharmacokinetics of Injected Insulin An optimal therapeutic use of insulin requires the knowledge of the factors affecting its absorption, disposal and action. Within 5–7 min, insulin given intravenously is concentrated in the heart, liver and kidneys and after 15 min mainly in the latter two organs. It has been shown that, in the range of physiological concentrations, liver extracts as much as 70% of insulin on a single passage, and that the kidney also removes a significant percentage of the insulin from the blood. The importance of liver and kidney in the insulin disposal is apparent as well as the need to adjust the insulin dosage in patients with hepatic or renal diseases. Insulin Concentration and Dose Insulin bioavailability is unaffected by insulin concentrations between 40 and 100 U/ml, while more diluted insulin is more rapidly absorbed. A more concentrated regular insulin (which has a more prolonged action) can be used for insulin-resistant patients. Increasing the dose of regular insulin delays the time of peak serum level and prolongs the duration of action, while increasing the dose of NPH insulin can reduce insulin absorption. It is noteworthy that, when the dose of insulin lispro is increased, the duration of action is not prolonged. Insulin Mixtures Manufactured insulin mixtures exist on the market. Biphasic premixed insulins have been developed in various ratios of rapid-acting to NPH (30/70, 40/60, 50/50, etc.). The effect of 30/70 mixtures of regular and NPH insulins is the same as if the components were injected separately and simultaneously, because regular insulin retains its pharmacokinetic characteristics. When a mixture of lente and regular insulins is used, the excess of zinc tends to bind to regular insulin and may cause precipitation of regular insulin out of solution, delaying its absorption and blunting its quick-acting effect. Thus, there are some advantages in using NPH for insulin mixtures. When two types of insulin are mixed, it is important to consider (to assure accuracy of the dose) the variable amount of ‘dead space’ between the hypodermic syringe and the needle. For this reason, it can be useful to always use syringes from the same manufacturer. Type of Administration and Site of Insulin Injection Subcutaneous administration (with all its disadvantages) remains the only practical method for the delivery of insulin. The peak concentration of insulin 76Belfiore/Iannello basic concept, it may be useful to bear in mind that the average insulin production in a normal subject is about 25 U/day. In diabetics, however, insulin requirement may be higher due to the presence of insulin resistance. Average starting doses of 0.7 U/kg body weight per day have been suggested. However, it may be more advisable to start with lower doses. In nonobese diabetics the starting daily insulin dose might be 15–20 U, whereas in obese diabetics it might be somewhat higher, in the range of 20–25 U. It is advisable to start therapy with caution, to avoid hypoglycemia, and to increase the insulin dose with prudence, after the previous dose has been experienced for a few days, with increases of 4–8 U per step. Insulin Regimens Insulin treatment can range from the simple regimens based on one or two daily insulin injections (conventional insulin therapy) to the more complicated multiple daily insulin injection regimens (intensive insulin therapy) or the continuous subcutaneous insulin infusion (CSII). Single Daily Insulin Injection It is the simpler and easier regimen to administer insulin. Some newly diagnosed type 1 diabetic patients (with some degree of residual b-cell function) can achieve glycemic control with less intensive effort by means of a single injection of lente or NPH insulin alone or combined with regular insulin. Type 2 diabetic patients, poorly controlled with diet and oral hypoglycemic drugs, can be treated with a single daily insulin injection (often a night injection of lente or NPH insulin will suffice), as well as older patients with impaired vision or physically disabled who may experience difficulties with injections. This regimen is also useful for individuals with limited motivation and poor compliance to the diabetic therapy. The initial insulin requirement can depend on several factors: (a) the current degree of hyperglycemia; (b) the dietary habit; (c) the amount of remaining endogenous insulin secretion; (d) the physical activity or exercise, and (e) the body weight or degree of obesity. A total dose of intermediate-acting insulin of 0.5–0.7 units/kg has been suggested (usually from 20–30 to 40–50 units/day), with subsequent adjust- ments according to the glycemic values obtained 2–4 times daily (as well as according to the hypoglycemic episodes, the presence of urinary ketones and the HbA1c values). With large doses of intermediate-acting insulin, availability of insulin may be inappropriate at certain times and there is the risk of nocturnal hypoglycemia. For some patients, an insulin preparation consisting 79Insulin Treatment in Type 1 and Type 2 Diabetes of a 30/70 mixture of regular and intermediate insulin can be used. In patients with persistent hyperglycemia there is the indication for changing insulin regi- men (twice-daily insulin injections – see below). When a hospitalized patient is discharged, the diet may be slightly increased or the insulin dose reduced because the physical activity of the patient will increase; for this reason, during the first days after discharge a contact between the patient and the physician is important to adjust the insulin dose. Twice-Daily Insulin Injection When a single dose is inadequate or produces hypoglycemia, twice-daily injections of NPH insulin are frequently used in insulin-dependent diabetics. Most patients should use two daily injections of a mixture of intermediate- acting (2/3) and rapid-acting (1/3) human insulins before breakfast and dinner. It is a relatively simple regimen that provides a good insulin availability over a 24-hour period, even if it may induce late afternoon and nocturnal hypoglyce- mia with pre-breakfast hyperglycemia. Twice-daily insulin regimen is little flexible, inasmuch as insulin must be given at the same time every day and meal times must also be kept constant. Multiple Daily Insulin Injections (MDI ) The intensive treatment regimens (as opposed to the simpler conventional regimens, described above) are not suitable for everyone and should be adopted in the appropriate patients. Intensive insulin therapy should be encouraged in type 1 diabetes without residual insulin secretion and when twice-daily insulin injections are no longer adequate. Recently, this regimen was also proposed for type 2 diabetics (particularly the younger patients with a life expectancy of 10–15 years or more). The scope is to obtain a good glycemic control which may reduce the development of diabetic microangiopathy, as shown by the DCCT (Diabetes Control and Com- plications Trial) study. However, this insulin intensive regimen may favor weight gain. Moreover, it has been postulated in the past that enhanced insulinization may be associated with increased risk of mortality from cardio- vascular disease and it was suggested that chronic hyperinsulinemia may cancel the beneficial effects of the better glycemic control. However, more recent data (from the DCCT for type 1 diabetes and from UKPDS for type 2 patients – see chapter VI) allow us to exclude that intensive insulin therapy entails risk for macrovascular disease. An absolute indication for intensive therapy is pregnancy (see chapter XVIII on Managing Diabetes and Preg- nancy). An optimal glycemic control requires that insulin delivery simulates the normal pattern of insulin secretion, which consists of continuous ‘basal’ insulin 80Belfiore/Iannello secretion (throughout the day and night) and acute increases of insulin levels connected to ingestion of meals. This regimen improves diabetic control, re- duces excursions in glycemic levels and provides a good flexibility. Four differ- ent regimens may be used: (a) The simplest intensive regimen entails the use of three injections, regular and intermediate-acting insulin before breakfast, regular insulin before supper and intermediate-acting insulin at bedtime. This 3 times daily insulin dose regimen is useful in diabetic patients with frequent nocturnal hypoglyce- mia and pre-breakfast hyperglycemia. The primary disadvantage of this approach is that meal schedules must be fixed rather rigidly. (b) Regular insulin before each meal and intermediate-acting insulin at bedtime (4 daily insulin doses). This regimen provides the greatest flexibility because regular insulin can be adjusted to cover each meal, avoiding postpran- dial hyperglycemia. (c) Regular and intermediate-acting insulin before breakfast, regular insu- lin before lunch and supper, and intermediate-acting insulin at bedtime (4 daily insulin doses). (d) Regular insulin before each meal and ultralente insulin in the morning (to replace basal insulin secretion) or subdivided before breakfast and before supper (4 daily insulin doses). It is less preferable to the (b) regimen because ultralente presents unexpected small peaks 15–24 h after injection. Human insulin lispro is very appropriate for multiple injection therapy, especially in patients with marked postprandial hyperglycemia and nocturnal hypoglycemia or with a variable lifestyle. Patients on insulin lispro had significantly lower glucose levels following meals (however with the potentially unwanted result of a rise in preprandial glucose) and showed a reduction in the incidence of severe hypoglycemia by 30% (compared to regular human insulin). In patients treated with insulin lispro (compared to those treated with human regular insulin) there should be less need for snacks. The majority of patients on insulin lispro reported an improved quality of life. However, there are some ‘failures’ with this type of insulin, as a number of patients may appear unable to control their diabetes with insulin lispro. At present, insulin lispro should be used with caution in children under the age of 12 as well as in gestational diabetes or pregnancy, because of lack of experience. Other, far too complex, multiple-injection regimens have also been sug- gested. Certainly, the adherence to therapy is less likely to occur when the program of treatment is far too complicated. Some patients object to such frequent needle injections and ask for changing from this insulin regimen to a simpler program. Pen devices or jet injectors filled with insulin (that are easy to carry) make the multiple daily insulin regimens better accepted. 81Insulin Treatment in Type 1 and Type 2 Diabetes (c) Blood glucose self-monitoring is the most important advance in dia- betes care. It requires the use of devices or meters that read blood glucose testing strips. All the modern meters can store and recall obtained blood glucose readings. These glucose determinations provide an estimation of gly- cemic control at any given moment, from day to day, and may be especially useful for specific problems (hypoglycemia, acute illness, ketonuria, periods of unstable diabetes, etc.). Several factors may limit the use of this method, such as a low level of motivation, a poor accuracy of determination, technical errors, intellectual inability to use the glycemic results, low visual or physical abilities, lack of education, high costs, etc. For some diabetic patients, blood glucose self-monitoring is perceived as too difficult or intrusive into individual’s routine, while other patients who desire to improve their glycemic control may accept to perform blood glucose tests several times a day on a regular basis. In these motivated patients, it is very important to monitor their technical competence, to define the desired glycemic range to be achieved, and to provide all the appropriate technical instructions, including the comparison of meter- obtained results with laboratory values. Glycated Hemoglobin (HbA1c) The patient with diabetes should have a periodic determination of HbA1c because this measurement is the most objective method of glucose control measurement over a long period. HbA is glycated in an irreversible and non- enzymatic fashion, and the levels of HBA1c reflect the mean glycemia over the 2–3 months prior to the test. Serum Fructosamine Test This test has been suggested as a less difficult to perform and less costly alternative to HBA1c determination, with which shows a good correlation. This test measures the level of glycosylated proteins in the blood (mainly albumin), and reflects the mean glycemic control during a 2- to 3-week period. Its validity is uncertain when interfering substances (bilirubin, hemolysis, etc.) are present or serum albumin concentration is abnormal. The test accuracy can be improved by correcting the fructosamine result for variations in serum albumin. HBA1c, compared to fructosamine test, should be considered as the preferable test for monitoring diabetic control. 84Belfiore/Iannello Complications of Insulin Treatment The most important complication of insulin therapy is hypoglycemia, which is discussed in chapter VIII (Clinical Emergencies in Diabetes. 2: Hy- poglycemia). The other complications are listed below. Insulin Edema In poorly controlled diabetic patients, insulin therapy can result in a marked accumulation of fluid, with localized (periorbital, pretibial or presa- cral) or generalized edema. The causes are probably multiple (table 4). A dietary restriction of salt and a temporary use of diuretics can be recom- mended. Edema will most often subside within 3–5 days. Table 4. Causes of insulin edema ADH increase (ascribed to hypovolemia resulting from osmotic diuresis) Cessation of natriuretic effect of hyperglucagonemia Increased plasma volume and transcapillary escape of albumin (with reduced colloid osmotic pressure) Excessive infusion of isotonic saline Na retention (induced by excess of insulin infused or injected) Insulin Lipoatrophy It was a common complication prior to the introduction of monocompo- nent insulins, consisting of a loss of fat at the site of insulin injection or, occasionally, at distant sites. In 25% of lipoatrophic patients, local allergy coexists. Lipoatrophy is frequently observed in young children (50%) or in young women (20%), compared to male adults (5%). Lipoatrophy, moreover, may occur after repeated injections of other substances such as narcotics or GH preparations. Thus, atrophy might be the result of a repeated mechanical trauma, even if insulin impurities can stimulate immune factors or immune complex formation which lead to local release of lipolytic substances. These reactions occur without overt inflammation and were considered also second- ary to insulin degradation or aggregation products. Indeed, in biopsy specimens of lipoatrophic areas, antigen-antibody reactions were not seen. Local reactions to protamine (a constituent of insulin preparations) and to silicone oil (the lubrificant in disposable syringes) may play a role in some patients. Switching to purified or human insulins and rotating the site of injections result in improvement of skin alterations in 97% of lipoatrophic patients. Very few cases were reported with recombinant human insulins, and the reason why it 85Insulin Treatment in Type 1 and Type 2 Diabetes still occasionally occurs is unknown. Injecting the purified insulin at the edges and center of the affected atrophic area improves lipoatrophy (due to the lipogenic effect of insulin). Addition of dexamethasone to the insulin in the syringe (4 lg/U, total daily dose not exceeding 0.75 mg) has also been sug- gested. In a recent case of severe well-circumscribed lipoatrophy, good results were obtained by treating the area with a fatty acid mixture while the patient was instructed to avoid this area for insulin injection. Insulin Lipohypertrophy It consists of visible or palpable increase of localized subcutaneous fat (most prevailingly in the anterior or lateral part of thighs) at the site of insulin injection, sometimes coexisting with lipoatrophy. Repeated and prolonged use of the same site for insulin injection is a main determinant in the development of lipohypertrophy. Often the affected patients report that injection into lipohy- pertrophic areas is less painful, perhaps because the subcutaneous tissue tends to be fibrous. Lipohypertrophy is due to a possible growth factor effect of insulin on cellular elements of subcutaneous tissue, and may alter the absorp- tion rate of insulin, thus possibly affecting metabolic control. Prevalence rates of lipohypertrophy vary between 20–45% in type 1 and 3–6% in type 2 diabetic patients. Independent risk factors which contribute to the presence of lipohy- pertrophy are female sex, type 1 diabetes, higher BMI and missing rotation of insulin injections. The most severe cases of insulin lipohypertrophy can be treated with liposuction, but prevention is important, primarily by systemat- ized rotation of injection sites within the recommended areas. An important role is played by educational interventions to establish an organized rotation system for insulin injection sites, to self-recognize lipohypertrophy and to normalize the high BMI. Syndrome of Immunologic Insulin Resistance All patients who receive insulin develop circulating antibodies, whose production can be influenced by several factors (table 5). Patients never treated with exogenous insulin may have circulating insulin antibodies, probably in- volved in the autoimmune reactions of type 1 diabetes. The high level of insulin antibodies may function as a reservoir from which insulin may be released unpredictably (thus inducing delayed hypoglycemia), or may bind insulin (thus causing hyperglycemia), or may form immune complexes (thus sequestering insulin in the reticuloendothelial system or stimulating procoagulant activity and favoring diabetic complications). In diabetic patients, this syndrome may result in an excessive insulin requirement (100–200 U/day in adults and up to 2.5 U/kg in children). In the most severe cases, steroids even at high doses for 3–4 weeks should be used. It is noteworthy that the immunogenicity of insulin 86Belfiore/Iannello Suggested Reading Anderson JH, Brunelle R, Koivisto VA: Reduction of post-prandial hyperglycemia and frequency of hypoglycemia in IDDM patients on insulin analog treatment. Diabetes 1997;46:265–270. Campbell PJ, May ME: A practical guide to intensive insulin therapy. Am J Med Sci 1995;310:24–30. Diabetes Control and Complications Trial Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–986. Dimitriadis G, Gerich J: Importance of timing of preprandial subcutaneous insulin administration in the management of diabetes mellitus. Diabetes Care 1983;6:374–377. Galloway JA, deShazo RD: Insulin chemistry and pharmacology; insulin allergy, resistance, and lipodys- trophy; in Rifkin H, Porte D (eds): Diabetes mellitus. Theory and Practice, ed 4. Amsterdam, Elsevier, 1990, pp 497–512. Sane T, Helve E, Yki-Jarvinen H: One year’s response to evening insulin therapy in non-insulin-dependent diabetes. J Intern Med 1992;231:253–260. Strowing S, Raskin P: Insulin treatment and patient management; in Rifkin H, Porte D (eds): Diabetes mellitus. Theory and Practice, ed 4. Amsterdam, Elsevier, 1990, pp 514–525. F. Belfiore, Institute of Internal Medicine, University of Catania, Ospedale Garibaldi, I–95123 Catania (Italy) Tel. +39 095 330981, Fax +39 095 310899, E-Mail francesco.belfiore@iol.it 89Insulin Treatment in Type 1 and Type 2 Diabetes Chapter VI Belfiore F, Mogensen CE (eds): New Concepts in Diabetes and Its Treatment. Basel, Karger, 2000, pp 90–102 ............................ Overview of Diabetes Management: ‘Combined’ Treatment and Therapeutic Additions F. Belfiore, S. Iannello Institute of Internal Medicine, University of Catania, Ospedale Garibaldi, Catania, Italy Lessons from Recent Large Trials on Diabetes Treatment The Diabetes Control and Complication Trial (DCCT), a large multicenter study conducted on more than 1,400 type 1 diabetics (aged 12–39 years) for a period of 7–10 years, has established that close blood glucose control (even if complete normalization of glycemic level was not obtained) reduces the frequency of late diabetic complications. Patients were assigned randomly to either intensive insulin therapy (3 or more daily injections or insulin pump, glucose self-monitoring 4 or more times per day, and frequent contact with a diabetes health-care team) or conventional therapy (1 or 2 injections of insulin mixtures per day, less frequent monitoring and medical contacts). The target goals of therapy were markedly different. Compared to the conventional care group, the intensive care group showed lower glycated hemoglobin (by 1.5–2.0%) and mean glucose level (by 60–80 mg/dl), yet most of the intensive care patients group failed to achieve normal glycemic levels. However, intensive care reduced the development of retinopathy by 76% (and its progression by 54%), the risk of microalbuminuria by 39%, frank proteinuria by 54%, and clinical neuropathy by 60%. Major cardiovascular events were also reduced, although statistical significance was not reached, in any case excluding that intensive insulin therapy may entail risk for macrovascular complications. The correlation of mean blood glucose with the frequency of retinopathy progression was linear, suggesting that there is no threshold glycemic level at which complications occur, so that any degree of improvement in glycemic control exerts beneficial effects on the progression of complications. These 90 beneficial effects, however, were obtained at the expense of a more common weight gain and, especially, of an increased (3-fold) risk of severe hypoglycemic episodes, often not accompanied by the classical symptoms (intensive treat- ment reduces the adrenergic response to hypoglycemia), which makes intensive treatment less appropriate for some people (hypoglycemia unawareness, special occupations, children, old people, etc.). Finally, it should be noted that the DCCT results were obtained through a close cooperation between the patients themselves and an expert team, primarily nurse educators and dieticians. There- fore, it may not be easy to follow the DCCT criteria in everyday clinical practice. The data from DCCT conclusively demonstrate that in type 1 diabetes the control of blood glucose really matters to prevent late complications. A recently concluded multicenter investigation on a very large study population (?5,000 patients), the United Kingdom Prospective Diabetes Study (UKPDS), whose results were presented at the European Association for the Study of Diabetes in Barcelona, September 1998, has obtained similar results in type 2 diabetic patients. As summarized by Laakso [1999], this study has shown that, compared to ‘diet alone’, the intensive control of blood glucose (regardless of the treatment used – sulfonylureas, metformin or insulin) reduced retinopa- thy or nephropathy by 25%, myocardial infarction by 16% and any diabetes- related endpoint by 12%. For every one percentage point reduction in HbA1c, there is a 35% reduction in retinopathy, nephropathy or neuropathy, and a 25% reduction in diabetes-related deaths (stroke frequency was not affected). As observed in the DCCT, there was no evidence of any glycemic threshold for micro- or macrovascular complications. With strict metabolic control, the risk of hypoglycemic episodes increased. Obese type 2 diabetic patients treated with metformin, compared with diet treatment, had even more pronounced benefits, showing reduction of 32 and 42% of diabetes-related endpoints and diabetes-related deaths, respectively, as well as a 36% reduction of all-cause mortality. In addition, they gained less weight and had fewer hypoglycemic episodes compared to the insulin- or sulfonylurea-treated patients. The UKPDS also pointed out that type 2 diabetic patients with tight control of blood pressure (mean 144/82 mm Hg), obtained either by ACE inhibitors or b-blockers, compared to the untreated group (154/87 mm Hg), showed reduction of any diabetes-related endpoint (by 24%), diabetes-related deaths (by 32%), stroke (by 44%) and microvascular complications (by 37%). Reduction of myocardial infarction of 21% occurred but did not reach statis- tical significance. It should be noted that in the UKPDS the treatment goal of maintaining fasting glycemia below 6 mmol/l (108 mg/dl) was not achieved. Strict metabolic control would consist of keeping glycemia below 10 mmol/l or 180 mg/dl at 91Overview of Diabetes Management For the patients who show secondary failure of sulfonylureas or respond only partially to maximum doses of oral sulfonylureas, various combined therapies can be employed. In the IDF guidelines [1999] to type 2 diabetes, the following combination therapy is suggested: (a) metformin with sulfonylu- reas; (b) sulfonylureas with a-glucosidase inhibitors, and (c) sulfonylureas with PPARc agonists. Combined Sulfonylurea-Metformin Therapy The oral hypoglycemic drugs, sulfonylureas and metformin, are largely used in combination in the treatment of type 2 diabetic patients, inasmuch as they exert different and complementary effects. Sulfonylureas stimulate insulin secretion whereas metformin ameliorates insulin action by enhancing periph- eral glucose utilization and repressing hepatic glucose production. Recent in vitro studies reported that metformin may potentiate glucose- stimulated insulin release from human pancreatic islets. Not enough evidence exists to support the suggestion that the association sulfonylureas-metformin entails a risk for diabetes-related deaths. In some countries, tablets containing a mixture of metformin and sulfonylureas (examples: 400 or 500 mg of metfor- min and 2.5 mg of glibenclamide) are available on the market. Combined Sulfonylurea-Insulin Therapy This combined therapy may be successful in type 2 insulin-resistant dia- betic patients who are no longer responsive to oral drugs. Sulfonylureas de- crease the exogenous insulin doses required to achieve a good glycemic control. Many studies have tested this therapeutic combination and have shown that about 30–40% of patients require significantly less insulin or sulfonylureas when treated this way. These patients, usually, show higher basal and stimulated serum C-peptide levels and increased insulin-mediated glucose disposal. The beneficial effects of the combination sulfonylurea-insulin can depend on an increase of endogenous insulin secretion or on a reduction of liver and periph- eral insulin resistance, and may be transient or prolonged. The recommended regimen is a dose of intermediate-acting insulin at night (to control overnight glucose production) and oral sulfonylureas during the day at meals (to reduce postprandial hyperglycemia). Therapy with Other Drugs Repaglinide. This is a nonsulfonylurea insulinotropic hypoglycemic agent of the meglitinide family (meglitinide is a compound with a poorly efficient insulinotropic activity), and shows a common conformation with the hypogly- cemic sulfonylureas glibenclamide and glimepiride (see chapter II on Insulin Secretion and Its Pharmacological Stimulation). 94Belfiore/Iannello Orlistat. In obese-diabetic patients who need to lose weight, a new nonsys- temically acting antiobesity drug, orlistat, may be a useful add-on. It possesses an inhibitory activity against gastrointestinal lipase A, thus selectively reducing the absorption of dietary fat in the gastrointestinal tract. After drug with- drawal, the lipase activity is rapidly restored, due to the continuous enzyme secretion. Orlistat has little or no effect on gastrointestinal enzymes other than lipase A such as amylase, trypsin, chymotrypsin and phospholipases. About 30% of dietary triglycerides remain undigested and is not absorbed, producing an additional caloric deficit compared to diet alone. Orlistat treatment also decreases the solubility and subsequent absorption of cholesterol, so improving lipid levels (both total and LDL cholesterol levels are reduced). More than 4,800 patients have received orlistat in clinical trials (the recommended dosage is 120 mg t.i.d. taken during meals), and the results demonstrate the efficacy (weight loss was 70% greater than with placebo plus diet), safety and toler- ability of the drug for long-term use. Orlistat-treated obese-diabetic patients present a best compliance with dietary restriction (because a severe dietary fat restriction is unnecessary) and a metabolic improvement (lowering of fasting blood glucose or HbA1c and reduction of sulfonylurea dosage requirement). This drug is free of systemic side effects, and gastrointestinal symptoms (related to the increased fecal fat excretion) are mild and self-limited. Orlistat treatment does not seem to increase the risk of gallstone formation (which can be favored by weight loss). Other Drugs. The possible use of some thiazolidinedione derivatives, such as pioglitazone, troglitazone, and rosiglitazone, has been discussed in chap- ter III (Insulin Resistance and Its Relevance to Treatment). Since fasting hyperglycemia in diabetes is correlated with high hepatic glucose production, which is determined by an elevated gluconeogenesis favored by FFA, inhibition of both FFA release (from adipose tissue) and oxidation (in the liver) may be an efficient modality to treat fasting hyperglycemia. Several drugs have been developed which inhibit FFA release from adipose tissue (acipimox, a nicotinic acid derivative) or hepatic FFA oxidation (etomoxir, a mitochondrial inhibitor of the carnitine palmitoyl transferase-1, the rate-limiting step in FFA oxida- tion). Severe fasting hypoglycemia and other side effects may occur with these drugs and limit their clinical use. Somatostatin or somatostatin analogues improve glucose metabolism in diabetic patients, especially under stress, selectively inhibiting the secretion of glucagon and GH without influencing insulin secretion. A role of somatostatin was also suggested in late diabetic vascular complications, but it remains to be elucidated. Amylin is a recently discovered 37-amino-acid peptide, that is cosecreted with insulin by pancreatic b-cells (it is regarded as the second b-cell hormone) 95Overview of Diabetes Management Table 1. Physiological actions of amylin or pramlintide Inhibition of food intake (through a central mechanism) Slowing of gastric motility and inhibition of gastric emptying Inhibition of postprandial glucagon secretion Suppression of postprandial hepatic glucose production Suppression of arginine-stimulated glucagon secretion Preservation of the glucagon increase in response to hypoglycemia Inhibition of insulin secretion in response to a variety of secretagogues Renal effects (stimulation of the renin-angiotensin-aldosterone system in rats and humans with possible induction of hypertension) Inhibition of gastric acid secretion, and gastroprotection in response to nutrient stimuli. It has been isolated and characterized as the major component of pancreatic amyloid deposits present in type 2 diabetic patients. In normal humans, plasma amylin concentrations vary in response to blood glucose levels, whereas in type 1 diabetic subjects and in late-stage type 2 diabetics it is reduced (being often almost undetectable) and do not increase in response to glucose load. Amylin secretion appears to be delayed and diminished in these populations. Human amylin tends to aggregate or forms insoluble particles and is not suitable for therapeutical use. A synthetic analog of human amylin, pramlintide, was developed, which is readily soluble in water and which possesses the same biological activities as amylin. The physiologic actions of human amylin or pramlintide are shown in table 1. Amylin appears to complement the glucose disposal actions of insulin and improves glucose regulation in type 1 or type 2 diabetic patients, who are absolutely or relatively deficient in amylin. This peptide, administered subcuta- neously at 10–100 lg 4 times/day, was able in a multicenter trial to effectively reduce the 24-hour plasma glucose profile in type 1 diabetic patients, without important side effects (only transient, dose-related, upper gastrointestinal symptoms such as nausea were observed). In type 2 diabetics treated with exogenous insulin, pramlintide improved metabolic control and produced sta- tistically significant reduction of serum fructosamine, HbA1c and total choles- terol as well as a trend towards decreased body weight. Type 1 Diabetes The treatment of type 1 diabetes with diet and insulin has been discussed in chapter IV on Diet and Modification of Nutrient Absorption and chapter 96Belfiore/Iannello