Type 1 diabetes	Type 2 diabetes
β-cell destruction	No β-cell destruction
Islet cell antibodies present	No islet cell antibodies present
Strong genetic link	Very strong genetic link
Age of onset usually below 30	Age of onset usually above 40
Faster onset of symptoms	Slower onset of symptoms
Insulin must be administered	Diet control and oral hypoglycaemic agents often sufficient control
Patients usually not overweight	Patients usually overweight
Extreme hyperglycaemia causes diabetic ketoacidosis	Extreme hyperglycaemia causes hyperosmolar hyperglycaemic state

Two other varieties of non-typical diabetes that may be seen are latent autoimmune diabetes in adults (LADA) and maturity-onset diabetes of the young (MODY). LADA occurs in younger, leaner individuals who appear to have type 2 diabetes as they do not become ketotic and may manage without insulin for a time. Antiglutamic acid decarboxylase (GAD) antibodies may be present and the individual usually progresses to insulin more rapidly than those with other varieties of type 2 diabetes. MODY was noted over 30 years ago and described a subset of type 2 diabetes of young onset, often with a positive family history. Genetic studies have now identified this to be a monogenic autosomal dominant form of diabetes. MODY related to the glucokinase gene typically causes a resetting of the glucose level with a ‘mild’ non-progressive hyperglycaemia in which diet treatment is usually sufficient. Other types of MODY are related to mutations in the hepatocyte nuclear factor genes and usually develop during adolescence or the early 20s. Pharmacological treatment is required, but sulphonylureas are extremely effective and insulin can usually be avoided.

Epidemiology

The incidence of type 1 diabetes is increasing worldwide, for unknown reasons. It is speculated that environmental changes may be causing modification to the diabetes-associated alleles. Also, since the introduction of insulin in the 1930s, an increasing number of people with type 1 diabetes have had children. There are major ethnic and geographical differences in the prevalence and incidence of type 1 diabetes. Figures are highest in Caucasians (especially Scandinavians), while the disorder is rare in Japan and the Pacific area. In northern Europe, the prevalence is approximately 0.3% in those under 30 years of age. Type 1 diabetes may present at any age, but there is a sharp increase around the time of puberty and a decline thereafter. Approximately 50–60% of patients with type 1 will present before 20 years of age.

Type 2 diabetes is much more common than type 1, accounting for 90% of people with diabetes. It usually occurs in those over the age of 40 years. Estimates in the UK suggest that type 2 diabetes currently affects approximately 2.3 million people, and up to another 500,000 are thought to be undiagnosed. The incidence of type 2 rises with age and with increasing obesity. As with type 1, there are major ethnic and geographical variations. In general, in non-obese populations, the prevalence is 1–3%. In the more obese societies, there is a sharp increase in prevalence with estimates of 6–8% in the USA, increasing to values as high as 50% in the Pima Indians of Arizona. Diabetes is six times more common among Asian immigrants in the UK than in the indigenous population. World studies of immigrants have suggested that the chances of developing type 2 are between two and 20 times higher in well-fed populations than in lean populations of the same race.

Aetiology

Both genetic and environmental factors are relevant in the development of type 1 diabetes, but the exact relationship between the two is still unknown. There is a strong immunological component to type 1 and a clear association with many organ-specific autoimmune diseases. Circulating islet cell antibodies (ICAs) are present in more than 70% of those with type 1 at the time of diagnosis. Family studies have shown that the appearance of ICAs often precedes the onset of clinical diabetes by as much as 3 years. Type 1 has been widely believed to be a disease of clinically rapid onset, but the development is related to a slow process of progressive immunological damage. However, it is not currently possible to use screening methods to reliably identify patients who will develop diabetes in the future. The final event that precipitates clinical diabetes may be caused by sudden stress such as an infection when the mass of β-cells in the pancreas falls below 5–10%.

Studies have been carried out in which patients with newly diagnosed type 1 were treated with immunosuppressive therapies such as ciclosporin, azathioprine, prednisolone and antithymocyte globulin. When started soon after diagnosis, these therapies showed transient improvements in clinical measures and increased the rate of remissions in which insulin was not required. However, their use is limited in an otherwise healthy and young population due to potential toxicity and the risks associated with immune suppression.

Studies have investigated the use of anti-CD3 monoclonal antibodies. When newly diagnosed type 1 patients are treated with short courses of anti-CD3 monoclonal antibodies, smaller insulin doses are required. This relates to better preservation of β-cell function.

Type 2 diabetes also has a strong genetic predisposition. Identical twins have a concordance rate approaching 100%, suggesting the relative importance of inheritance over environment. If a parent has type 2, the risk of a child eventually developing type 2 is 5–10% compared with 1–2% for type 1. Type 2 diabetes occurs because of the progressive development of insulin resistance and β-cell dysfunction, the latter leading to an inability of the pancreas to produce enough insulin to overcome the insulin resistance. About 85% of people with type 2 diabetes are obese. This highlights the clear association between type 2 and obesity, with obesity causing insulin resistance. In particular, central obesity, where adipose tissue is deposited intra-abdominally rather than subcutaneously, is associated with the highest risk. Body mass index (BMI) has been used as an indicator for predicting type 2 risk; however, it does not take fat distribution into account, so waist circumference measurements are now being increasingly used.

Pathophysiology

The islets of Langerhans form the endocrine component of the pancreas, constituting 1% of the total pancreatic mass. Insulin is synthesised in the pancreatic β-cells, initially as a polypeptide precursor, preproinsulin. The latter is rapidly converted in the pancreas to proinsulin. This forms equal amounts of insulin and C-peptide through removal of four amino acid residues. Insulin consists of 51 amino acids in two chains (the A chain contains 21 amino acids and B chain contains 30), connected by two disulphide bridges. In the islets, insulin and C-peptide (and some proinsulin) are packaged into granules. Insulin associates spontaneously into a hexamer containing two zinc ions and one calcium ion.

Glucose is the major stimulant to insulin release. The response is triggered both by the intake of nutrients and the release of gastro-intestinal peptide hormones. Following an intravenous injection of glucose, there is a biphasic insulin response. There is an initial rapid response in the first 2 min, followed after 5–10 min by a second response which is smaller but sustained over 1 h. The initial response represents the release of stored insulin and the second phase reflects discharge of newly synthesised insulin. Glucose is unique; other agents, including sulphonylureas, do not result in insulin biosynthesis, only release. Once released from the pancreas, insulin enters the portal circulation. The liver rapidly degrades it and only 50% reaches the peripheral circulation. In the basal state, insulin secretion is at a rate of approximately 1 unit/h. The intake of food results in a prompt five- to tenfold increase. Total daily secretion is approximately 40 units.

Insulin circulates free as a monomer, has a half-life of 3–5 min and is primarily metabolised by the liver and kidneys. In the kidneys, insulin is filtered by the glomeruli and reabsorbed by the tubules and degraded. In both renal and hepatic disease, there is a decrease in the rate of insulin clearance, which may necessitate dosage reduction for those using exogenous insulin. Peripheral tissues such as muscle and fat also degrade insulin, but this is of minor quantitative significance.

The interaction of insulin with the receptor on the cell surface sets off a chain of messengers within the cell. This opens up transport processes for glucose, amino acids and electrolytes.

In type 1 diabetes, there is an acute deficiency of insulin that leads to unrestrained hepatic glycogenolysis and gluconeogenesis with a consequent increase in hepatic glucose output. Also, glucose uptake is decreased in insulin-sensitive tissues such as adipose tissue and muscle; hence, hyperglycaemia ensues. Either as a result of the metabolic disturbance itself or secondary to infection or other acute illness, there is increased secretion of the counter-regulatory hormones glucagon, cortisol, catecholamine and growth hormone. All of these will further increase hepatic glucose production.

In type 2 diabetes, the process is usually less acute, since insulin production decreases over a sustained period of time. Hyperinsulinaemia is able to maintain glucose levels for a period of time, but eventually β-cell function deteriorates and hyperglycaemia ensues. If this cycle is not interrupted, type 2 diabetes develops. Impaired glucose tolerance (IGT), impaired fasting glucose or hyperinsulinaemia may be detected before overt diabetes develops, and if so, a strict diet and exercise regimen leading to weight loss and improved insulin sensitivity may delay or even prevent the onset of diabetes. At the time of diagnosis, those with type 2 diabetes may have already lost about 50% of their β-cell function. Irrespective of treatment, β-cell function continues to decline with time, often leading to the need for regular insulin therapy.

Type 2 diabetes is also associated with the metabolic syndrome (or syndrome X), although the real relevance of this ‘syndrome’ continues to be debated in the literature (Khan et al., 2005). The metabolic syndrome is a group of risk factors commonly found in those with type 2 diabetes, including insulin resistance, glucose intolerance (type 2 diabetes or IGT), hyperinsulinaemia, hypertension, dyslipidaemia, central obesity, atherosclerosis and increased levels of procoagulant factors, for example, plasminogen activator inhibitor-1 and fibrinogen.

Pathophysiology of insulin resistance

Abdominal fat, found in abundance in the majority of those with type 2 diabetes, is metabolically different from subcutaneous fat and can cause ‘lipotoxicity’. Abdominal fat is resistant to the antilipolytic effects of insulin, resulting in the release of excessive amounts of free fatty acids, which in turn lead to insulin resistance in the liver and muscle. The effect is an increase in gluconeogenesis in the liver and an inhibition of insulin-mediated glucose uptake in the muscle. Both these result in increased levels of circulating glucose. Further, excess fat itself may contribute to insulin resistance because when adipocytes become too large they are unable to store additional fat, resulting in fat storage in the muscles, liver and pancreas, causing insulin resistance in these organs.

Excess intra-cavity adipose tissue causes the oversecretion of some cytokines (adipokines or adipocytokines) associated with inflammation, endothelial dysfunction and thrombosis. Examples of such adipokines include plasminogen activator inhibitor-1 (which is prothrombotic), tumour necrosis factor-α and interleukin-6 (which are proinflammatory) and resistin (which causes insulin resistance). The atherosclerosis associated with insulin resistance is due to hypercoagulability, impaired fibrinolysis and the toxic combination of endothelial damage (caused by chronic, subclinical inflammation), oxidative stress and hyperglycaemia. Excess adipose tissue is also thought to cause undersecretion of a beneficial adipokine called adiponectin. Adiponectin suppresses the attachment of monocytes to endothelial cells, thereby protecting against vascular damage. People with type 2 diabetes have lower levels of adiponectin than those without diabetes and weight reduction increases adiponectin levels.

Clinical manifestations

The symptoms of both type 1 and type 2 diabetes are similar, but they usually vary in intensity. Those associated with type 1 diabetes are more severe and faster in onset. The symptoms are related to the osmotic effects of glucose and the abnormalities of energy partitioning. Common symptoms include polyuria (increased urine production, particularly noticeable at night) and polydipsia (increased thirst). These are a consequence of osmotic diuresis secondary to hyperglycaemia. These symptoms are frequently accompanied by fatigue due to an inability to utilise glucose and marked weight loss because of the breakdown of body protein and fat as an alternative energy source to glucose. Blurred vision caused by a change in lens refraction may also occur and patients should be advised that as glucose levels are normalised, vision normally improves and new spectacles should be avoided for the first 3 months of effective treatment of the hyperglycaemia. Patients may also experience a higher infection rate, especially Candida, and urinary tract infections due to increased urinary glucose levels.

Type 1 diabetes

The metabolic abnormality at presentation of an individual with type 1 diabetes is often profound. The symptoms mentioned earlier are usually extreme and of recent (days or weeks) onset. In a significant proportion, the presenting symptoms are those of diabetic ketoacidosis, nausea, vomiting, dehydration, shortness of breath secondary to an attempt by the respiratory system to neutralise the metabolic acidosis caused by the ketones and, in extreme cases, coma.

Type 2 diabetes

Many patients with type 2 diabetes have an insidious onset of hyperglycaemia, with few or no classic symptoms. This is particularly true in obese individuals, whose diabetes may be detected only after glycosuria or hyperglycaemia is found serendipitously. Some patients are unaware of the disease even with marked classic symptoms as they begin so gradually and over such a long period of time. Recurring infections, for example, urinary tract, chest, soft tissue, are common because sustained hyperglycaemia can result in severe impairment of phagocyte function, and raised glucose levels provide a growth medium for bacteria.

Generalised pruritus and symptoms of vaginitis, which may be due to candidal infection, are frequently the initial complaints of women with type 2 diabetes. Patients often present when the complications of sustained hyperglycaemia have already developed, for example, cardiovascular disease or renal disease. Retinopathy may be detected on routine ophthalmological examination. Alternatively, a combination of neuropathy, peripheral vascular disease (PVD) and infection may manifest as foot ulceration or gangrene. In some cases, patients present with hyperosmolar hyperglycaemic state (HHS) where glucose levels in excess of 35 mmol/L are found and excessive dehydration has occurred. Occasionally, patients with type 2 diabetes present with diabetic ketoacidosis, especially in severe infection or in those of African/Caribbean descent.

Diagnosis

In June 2000, the UK formally adopted the World Health Organization criteria for diagnosing diabetes mellitus that was initially published in 1999. It has since been updated and the diagnostic criteria have been reiterated (World Health Organization, 2006).

1. Diabetes symptoms (i.e. polyuria, polydipsia and unexplained weight loss) plus:

• a fasting serum glucose concentration ≥7.0 mmol/L

• or serum glucose concentration ≥11.1 mmol/L 2 h after 75 g anhydrous glucose in an oral glucose tolerance test (see later).

2. With no symptoms, diagnosis should not be based on a single glucose determination but requires confirmatory serum venous determination. At least one additional glucose test result, on another day with the value in the diabetic range, is essential, either fasting or from the 2-h post-glucose load. If the fasting value is not diagnostic, the 2-h value should be used.

Current recommendations are that the diagnosis is confirmed by a glucose measurement performed in an accredited laboratory on a venous serum sample. A diagnosis should never be made on the basis of glycosuria or a stick reading of a finger prick blood glucose alone, although such tests are being examined for screening purposes. Glycated haemoglobin (HbA_1c) is also not currently recommended for diagnostic purposes, although this is currently being considered.

Glucose tolerance test

The patient should not be taking any drugs which interfere with glucose handling. A normal diet with at least 150 g of carbohydrate per day should be consumed for the 3 days before the test, but the patient should be fasted from 8 pm (except for water) on the day before the test. The test should commence at around 9 am with a venous serum glucose test, followed by the administration of 75 g of glucose by mouth over a 5-min period. This is often given in the form of 394 mL of Lucozade® original. The second venous serum glucose sample is then taken 2 h after the drink. The patient should be seated and is not permitted to smoke, eat or drink anything other than water until the test is complete. As there is a risk of later-onset hypoglycaemia in some individuals, it is advisable to suggest that the patient has something to eat immediately upon completion of the test, especially if he/she is planning to drive.

Diabetic emergencies

Hypoglycaemia and extreme hyperglycaemia, causing diabetic ketoacidosis or hyperosmolar hyperglycaemic state, constitute the three acute emergencies associated with diabetes.

Hypoglycaemia

Hypoglycaemia can occur both with insulin treatment and in those taking some oral agents, especially the longer-acting sulphonylureas, for example, chlorpropamide and glibenclamide. Definitions of hypoglycaemia vary, and in particular, there is no WHO definition. However, symptoms caused by the release of counter-regulatory hormones predominantly adrenaline (epinephrine), noradrenaline (norepinephrine) and glucagon tend to occur when the venous serum glucose drops below 3.0 mmol/L in healthy individuals. These symptoms described in Box 44.2 are a normal physiological response to hypoglycaemia and should alert the person to consume carbohydrates. Individuals may not respond appropriately to hypoglycaemia of this degree for several reasons, termed hypoglycaemia unawareness. First, the relevance of the symptoms has not been explained to them. This is an educational failing. It is imperative, therefore, that people with diabetes who are prescribed medication which is known to cause hypoglycaemia should be educated about the autonomic symptoms so that they may take action to avoid further decline of serum glucose. Second, the symptoms simply may not occur because of autonomic neuropathy. One of the commonest complications of diabetes is neuropathy, and when this includes the autonomic nervous system, there are no reliable symptoms to warn the individual that they are hypoglycaemic. A similar situation may occur as a consequence of drugs which suppress autonomic symptoms, such as β-blockers. Third, the patient may have recurrent hypoglycaemia. In those individuals who suffer frequent hypoglycaemic episodes, the autonomic symptoms may cease to occur. There is some evidence that the symptoms can be regained if, for a period of a few weeks, the serum glucose level can be maintained out of the hypoglycaemic range. Finally, the individual may be hypoglycaemia unaware because of alcohol intoxication.

If the serum glucose is allowed to drop to around 2 mmol/L, there are acute changes in cerebral function which lead initially to confusion. This is followed by coma, seizures and death if the glucose drops below about 0.5 mmol/L. Any cerebral malfunction is termed neuroglycopaenia.

Causes of hypoglycaemia

The most common causes of hypoglycaemia are either a decrease in carbohydrate consumption, excess carbohydrate utilisation from unexpected exercise or increase in circulating insulin (Table 44.2).

Table 44.2 Causes of hypoglycaemia

Cause	Comment
Missed meals or delays in eating	Reduced carbohydrate intake, therefore reduction in glucose levels
Not eating the usual amount of carbohydrates	Reduced carbohydrate intake, therefore reduction in glucose levels
Increased doses of insulin	Increased uptake of glucose into cells and increased storage of glucose as glycogen
Increased doses of oral insulin secretagogues	Increased levels of insulin therefore increased uptake of glucose into cells and increased storage of glucose as glycogen
Introduction of other blood glucose-lowering agents to oral insulin secretagogues	Enhanced hypoglycaemic effects
Increase in exercise	Increased uptake of glucose into cells
Excessive alcohol consumption	Impaired gluconeogenesis
Liver disease	Impaired gluconeogenesis and glycogenolysis

Nocturnal hypoglycaemia

Sometimes, hypoglycaemia occurs throughout the night. Symptoms may include restlessness, although this may not be identified unless observed by another person. When nocturnal hypoglycaemia occurs, the person often wakes feeling unrested, unwell or with a headache. Contrary to what might be expected, morning blood glucose readings may be high because a sustained hypoglycaemic episode leads counter-regulatory hormones to raise blood glucose levels. This could present a confusing picture as the obvious solution to a raised blood glucose level in the morning would be to increase the evening/night-time dose of insulin. However, in the case of nocturnal hypoglycaemia, this would make the problem worse. If nocturnal hypoglycaemia is suspected, then blood glucose should be measured at night, for example, 2.00–3.00 am. If confirmed, the patient should either have a snack before bedtime, reduce the evening/night-time dose of insulin, alter the timing of administration of the evening dose of intermediate- or long-acting insulin in order to delay the peak of bioavailability or change the intermediate-acting insulin to a peakless analogue as appropriate. It is important to discuss nocturnal hypoglycaemia with patients as there is often a fear of dying from unrecognised hypoglycaemia in sleep. This fear is unfounded because of the protection from hypoglycaemia severe enough to cause death afforded by the counter-regulatory hormones. This occurs even in those individuals with autonomic neuropathy.

Treatment of hypoglycaemia

If the patient is able to swallow safely without the risk of aspiration, then glucose should be taken orally. However, if unable to swallow or if there is a risk of aspiration because, for example, of a decreased level of consciousness, parenteral treatment should be given, either intravenous glucose or intramuscular glucagon.

The most effective oral treatments are pure sources of glucose, for example, five glucose tablets or glucose drinks such as 150 mL of Lucozade®. In an emergency, hot drinks should be avoided as they might burn and drinks containing milk are not suitable as the fat in milk slows down sugar absorption. Blood glucose levels should be measured about 10–15 min after treating hypoglycaemia. If below 3.5 mmol/L, more glucose should be consumed. If above 3.5 mmol/L and the next meal will be over 1 h, then a long-acting carbohydrate is also required, for example, bread or biscuits. However, if the person is taking an α-glucosidase inhibitor such as acarbose, then monosaccharide carbohydrates must be given because disaccharides and polysaccharides will not be absorbed due to inhibition of the enzymes cleaving carbohydrate into absorbable monosaccharide units.

Should parenteral treatment be required, 25 g of intravenous glucose or 1 mg of intramuscular glucagon is recommended. Glucagon takes approximately 15–20 min to work, but if the person has liver disease (cirrhosis) or is malnourished, then glucagon may not work because glucagon acts by mobilising glucose stores from the liver. In such cases, intravenous glucose must be given. A number of serious extravasation injuries, some necessitating amputation of the affected limb, have been caused by 50% glucose. As a consequence, many hospitals now use 20% glucose.

Diabetic ketoacidosis

Diabetic ketoacidosis is serious, and in developed countries, it has a mortality rate of 5–10%. It occurs because absence of insulin causes extreme hyperglycaemia. At the same time, the normal restraining effect of insulin on lipolysis is removed. Non-esterified fatty acids are released into the circulation and taken up by the liver, which produces acetyl coenzyme A (acetyl CoA). The capacity of the tricarboxylic acid cycle to metabolise acetyl CoA is rapidly exceeded. Ketone bodies, acetoacetate and hydroxybutyrate are formed in increased amounts and released into the circulation. Further, osmotic diuresis, caused by hyperglycaemia, lowers serum volume, causing dizziness and weakness due to postural hypotension. Weakness is increased by potassium loss, caused by urinary excretion and vomiting due to stimulation of the vomiting centre by ketones, and catabolism of muscle protein. When insulin deficiency is severe and of acute onset, all of these symptoms are accelerated. Ketoacidosis exacerbates the dehydration and hyperosmolarity by producing anorexia, nausea and vomiting. As serum osmolarity rises, impaired consciousness ensues with coma developing in approximately 10% of cases. Metabolic acidosis causes stimulation of the medullary respiratory centre, giving rise to Kussmaul respiration (deep and rapid breathing) in an attempt to correct the acidosis. The patient’s breath may have the fruity odour of acetone (ketones) commonly described as smelling like pear drops or nail varnish remover.

Precipitating factors for diabetic ketoacidosis in type 1 disease are usually omission of insulin dose, acute infection, trauma or myocardial infarction. Although diabetic ketoacidosis is normally associated with type 1 diabetes, it may rarely occur in people with type 2.

Diagnosis of diabetic ketoacidosis

Diagnosis requires demonstration of hyperglycaemia and metabolic acidosis with the presence of ketones. The biochemical diagnosis of ketoacidosis is usually made at the bedside and confirmed in the laboratory. Urinalysis will show marked glycosuria and ketonuria. A blood glucose test strip usually shows a blood glucose level of more than 22 mmol/L. Formal laboratory measurement of glucose, urea, creatinine, electrolytes and venous bicarbonate should be carried out. Two potentially misleading laboratory results are the white blood cell count and serum sodium. The former will always be raised but correlates with the ketone body level and is not therefore a guide to infection. The serum sodium level will often be low due to the osmotic effect of glucose draining water from the cells and diluting the sodium. The sodium concentration will also be spuriously low if there is marked dyslipidaemia.

Treatment of diabetic ketoacidosis

Treatment comprises fluid volume expansion (initially with 0.9% sodium chloride), correction of hyperglycaemia and the presence of ketones (by infusion of insulin), prevention of hypokalaemia, and identification and treatment of any associated infection. Once the patient is better, they should be reviewed by the diabetes team in order to discuss how to avoid future episodes of diabetic ketoacidosis.

Hyperosmolar hyperglycaemic state

HHS is associated with type 2 disease and has a higher mortality rate (15%) than diabetic ketoacidosis. HHS usually occurs in middle-aged or elderly people, about 25% of whom have previously undiagnosed type 2 diabetes.

In HHS, unlike diabetic ketoacidosis, there is no significant ketone production and therefore no severe acidosis. Hyperglycaemia occurs gradually over a sustained period of time, leading to dehydration due to osmotic diuresis which, if severe, results in hyperosmolarity. Hyperosmolarity may increase blood viscosity and the risk of thromboembolism. Factors precipitating HHS are infection, myocardial infarction, poor adherence with medication regimens or medicines which cause diuresis or impair glucose tolerance, for example, glucocorticoids.

Diagnosis of HHS

The diagnostic features of HHS are hyperglycaemia (often in the region of 55 mmol/L, which is generally much higher than for diabetic ketoacidosis), dehydration and hyperosmolarity. There may be a mild metabolic acidosis but without marked ketone production. Conscious levels on presentation range from slight confusion to coma. In some cases, seizures occur. Serum sodium and potassium levels are usually normal, but creatinine is high. The average fluid deficit is 10 L, so circulatory collapse is common.

Treatment of HHS

Treatment requires fluid replacement to stabilise blood pressure and improve circulation and urine output. Sodium chloride 0.9% or 0.45% (if serum sodium is greater than 150 mmol/L) is given and monitoring of blood pressure and cardiovascular status undertaken. Potassium may be added if required. Insulin treatment is started via intravenous infusion but is not aggressive, since fluid replacement also lowers serum glucose levels. Prophylaxis or treatment for thromboembolism may also be required.

Long-term diabetic complications

Diabetes and its long-term complications cost the NHS substantial amounts of money – approximately 10% of the total budget (£173 million/week).

Although all long-term complications may occur in each type of diabetes, the spectrum of incidence is different. Many patients with type 2 diabetes have had their disease a long time before the diagnosis, by which time many have developed diabetic complications (Figs. 44.1 and 44.2). However, diabetic complications can be limited and sometimes prevented altogether if good management occurs from an early stage. Hyperglycaemia and hypertension are the two major modifiable risk factors that influence the development of diabetic complications.

Fig. 44.1 Incidence of macrovascular complications at diagnosis in type 2 diabetes.

Fig. 44.2 Incidence of microvascular complications at diagnosis in type 2 diabetes.

Patients with diabetes should undergo regular review of their disease management for early signs of associated complications and review of these risk factors and their management. Diabetic complications are frequently divided into macrovascular and microvascular complications. Macrovascular complications arise from damage to large blood vessels and microvascular complications occur from damage to smaller vessels. The general aetiology of macro- and microvascular complications is the same and results from atherosclerosis of the vessels, which may lead to occlusion. The main aims of treatment are, first, to prevent the immediate symptoms associated with diabetes, for example, polyuria, polydipsia, etc., and second, to prevent development, or slow the progression of the long-term disease complications.

Macrovascular disease

The risk of macrovascular complications, including cardiovascular disease (coronary heart disease and stroke) and PVD, is 2–4 times higher for people with diabetes.

Microvascular disease

Microvascular complications include retinopathy, nephropathy and neuropathy.

Retinopathy

Diabetic retinopathy is the leading cause of blindness in people under the age of 60 in industrialised countries. Twenty years from the onset of diabetes, over 90% of people with type 1, and over 60% of people with type 2, will have diabetic retinopathy. The main problem with diagnosing retinopathy is that it is symptomless until the disease is far advanced. Therefore, if regular screening is not undertaken, diagnosis may not be made early enough for successful treatment intervention. Tight glycaemic control has been shown to prevent and delay the progression of retinopathy in patients with type 1 disease. Likewise, for patients with type 2 diabetes, both tight glycaemic control and tight blood pressure control reduce the risk of developing retinopathy. When retinopathy is detected early, sight may be saved by laser photocoagulation. In advanced cases, surgery may be required.

Pregnancy may worsen moderate-to-severe retinopathy, particularly if there is poor or sudden improvement in glycaemic control. However, tight glycaemic control during pregnancy reduces the long-term risks of retinopathy.

Nephropathy

In diabetic renal disease, the kidneys become enlarged and the glomerular filtration rate (GFR) initially increases. However, if the nephropathy progresses, the GFR starts to decline. Serum creatinine used alone to estimate renal function has limitations. The GFR can be estimated (eGFR). The most popular method is the modified Modification of Diet in Renal Disease (MDRD) formula, which requires serum creatinine, age, sex and ethnicity:

The presence of nephropathy is indicated by the detection of microalbuminuria (small amounts of albumin present in urine). If higher amounts of albumin are detected, this is termed proteinuria (or macroalbuminuria) and signifies more severe renal damage. Microalbuminuria is defined as an albumin:creatinine ratio (ACR) greater or equal to 2.5 mg/mmol (men) and 3.5 mg/mmol (women). Proteinuria may be defined as an albumin:creatinine ratio greater than 30 mg/mmol or albumin concentration greater than 200 mg/L. Proteinuria may progress to end-stage renal disease and require dialysis. Albumin in the urine increases the risk of cardiovascular disease, with microalbuminuria associated with 2–4 times the risk, proteinuria with nine times the risk and end-stage renal disease increasing risk by 50 times.

Tight control of both glycaemic levels and blood pressure reduces the risk of developing nephropathy. Angiotensin-converting enzyme (ACE) inhibitors and/or angiotensin receptor blockers (ARBs) are the treatments of choice, since both have been demonstrated to provide renal protective effects additional to their antihypertensive effects. ACE inhibitors and ARBs have been shown to delay the progression to proteinuria in patients with microalbuminuria. Although not proven for all individual drugs in these classes, it is considered to be a class effect. However, these drugs should be used with care if there is a risk of renovascular disease.

Peripheral neuropathy

Peripheral neuropathy is the progressive loss of peripheral nerve fibres resulting in nerve dysfunction. Diabetic neuropathies can lead to a wide variety of sensory, motor and autonomic symptoms. The most common is the symmetrical distal sensory type, which is particularly evident in the feet and may slowly progress to a complete loss of feeling. It is most prevalent in elderly patients with type 2 diabetes but may be found with any type of diabetes, at any age beyond childhood. Painful diabetic neuropathy is another manifestation of sensory neuropathy; it can be extremely disabling and may cause considerable morbidity. Guidance on the treatment of painful neuropathy is available (National Institute for Health and Clinical Excellence, 2010). Diabetic proximal motor neuropathy is rapid in onset and involves weakness and wasting, principally of the thigh muscles. Muscle pain is common and may require opiate analgesia. Distal motor neuropathy can lead to symptoms of impaired fine co-ordination of the hands and/or foot slapping.

Autonomic neuropathy may affect any part of the sympathetic or parasympathetic nervous systems. The most common manifestation is diabetic impotence. Bladder dysfunction usually manifests as loss of bladder tone with a large increase in volume. Diabetic diarrhoea is uncommon, but can be troublesome as it tends to occur at night. Gastroparesis may cause vomiting and delayed gastro-intestinal transit and variable food absorption, causing difficulty in the insulin-treated patient. Postural hypotension due to autonomic neuropathy is uncommon but can be severe and disabling. Disorders of the efferent and afferent nerves controlling cardiac and respiratory function are more common, but rarely symptomatic. Autonomic neuropathy may also cause dry skin and lack of sweating, both of which may contribute to diabetic foot problems.

Macro- and microvascular disease combined

Diabetic foot problems

Infected diabetic foot ulcers account for the largest number of diabetes-related hospital bed-days and are the most common non-trauma cause of amputations. The rate of lower-limb amputation in people with diabetes is 15 times higher than in the general population. The lifetime risk of a person with diabetes developing a foot ulcer is around 15%. Diabetic foot ulcers are a costly problem and are associated with considerable morbidity. Foot problems often develop as a result of a combination of specific problems associated with having diabetes, that is sensory and autonomic neuropathy, PVD and hyperglycaemia. Poor foot care and poorly controlled diabetes are also contributory factors. Development of foot ulcers may be partly preventable by patient education. People with diabetes learn that their feet are particularly vulnerable, and if problems arise, they must seek immediate professional advice.

There are three main types of foot ulcers: neuropathic, ischaemic and neuroischaemic. Neuropathic ulcers occur when peripheral neuropathy causes loss of pain sensation. The ulcers can be deep but are usually painless and are caused by trauma to the foot which is not noticed until after significant damage has occurred. Ischaemic ulcers result from PVD and poor blood supply causing a reduction in available nutrients and oxygen required for healing. Ischaemic ulcers are painful and usually occur on the distal ends of the toes. Most ulcers have elements of both neuropathy and ischaemia and are termed neuroischaemic.

Diabetic foot ulcers are prone to infection, the most common pathogens being staphylococci and streptococci. Wounds with an ischaemic component are commonly infected with anaerobic organisms.

Charcot arthropathy

This is an uncommon foot complication caused by severe neuropathy. It results in chronic, progressive destruction of joints with marked inflammation. Reduced bone density leads to bone fractures, altered foot shape and gross deformity. Owing to the deformity which occurs, excess pressure over protruding bone frequently leads to ulceration unless footwear is extensively modified.

Treatment

Treatment for people with diabetes includes advice on nutrition, physical activity, weight loss and smoking cessation if appropriate. Drug therapy is prescribed where necessary.

Structured education programmes

It is now considered standard care to offer people with diabetes education to help them manage their diabetes themselves on a day-to-day basis. This should be offered in a structured way and based upon theories of adult learning. Group education is now invariably offered as it offers economic advantages over one-to-one education as well as peer support. Examples of such programmes include DAFNE (Dose Adjustment For Normal Eating) in type 1 diabetes and DESMOND (Diabetes Education for Self Management in the Ongoing and Newly Diagnosed) in type 2 diabetes.

Diet

Dietary control is the mainstay of treatment for type 2 diabetes and plays an integral part in the management of type 1. Dietary recommendations have undergone extensive review in recent years and considerable changes have been made. Generally speaking, healthy eating advice for people with diabetes is the same as for the general population. Some of the general dietary advice that patients should be given is shown in Box 44.3.

Box 44.3 General dietary advice for people with diabetes

• Eat regular meals based on starchy foods such as bread, pasta, potatoes, rice and cereals. Whenever possible, choose high-fibre varieties of these foods, for example, wholemeal bread and wholemeal cereals which have a lower glycaemic index.

• Try to cut down on fat, particularly saturated (animal) fats. Monounsaturated fats such as olive oil are preferred. Use less butter, margarine, cheese and eat fewer fatty meals. Choose low-fat dairy foods, for example, skimmed milk and low-fat yoghurt. Grill, steam or oven bake instead of frying or cooking with oil or other fats.

• Try to eat at least five portions of fruit and vegetables every day. This provides vitamins and fibre as well as helping to balance the overall diet.

• Cut down on sugar and sugary foods. Sugar can still be used as an ingredient in foods and baking as part of a healthy diet. Use sugar-free, low-sugar or diet squashes and fizzy drinks, as sugary drinks cause blood glucose levels to rise quickly.

• Use less salt as high intake can raise blood pressure. Food can be flavoured with herbs and spices instead of salt.

• Drink alcohol in moderation. Two units/day for a woman and three for a man. A small glass of wine or half a pint of normal strength beer is one unit. Never drink on an empty stomach as alcohol can exacerbate hypoglycaemia.

Carbohydrates and sweeteners

The blood glucose level is closely affected by carbohydrate intake. Previous guidance for people with diabetes recommended eating about the same amount of carbohydrate at approximately the same time each day and generally advised restriction of carbohydrates. As a consequence of this advice, a number of people tended to eat more fat. Current guidance for carbohydrate consumption still emphasises the importance of total carbohydrate intake, but it focuses on selecting carbohydrates with a lower glycaemic index, that is carbohydrates which give sustained release of sugars over time, as opposed to carbohydrates with a high glycaemic index that give high peaks in blood glucose concentration. Examples of carbohydrates with a low glycaemic index include beans, pulses and starchy foods like wholemeal pasta and wholegrain bread. Total carbohydrate consumption should not exceed 45–60% of energy intake, with monounsaturated fat and carbohydrate combined making up 60–70% of energy intake.

Sucrose or ‘sugar’ may be included in the diet, according to the new guidance, but sucrose should account for no more than 10% of total energy and should be spaced throughout the day, rather than being consumed all in one go. Sugar alcohols, for example, sorbitol, maltitol and xylitol, often used as sugar substitutes in diabetic foods, are expensive and may cause diarrhoea. They are therefore considered to confer little advantage over sucrose. Non-nutritive or intense sweeteners such as aspartame, saccharin, acesulphame K, cyclamate and sucralose may be useful, especially for those who are overweight.

Alcohol

Alcohol contains carbohydrates and, if consumed in excess, may cause hyperglycaemia. However, more dangerously, it is also associated with later-onset (up to 16 h post-alcohol) hypoglycaemia and hypoglycaemia unawareness. Alcohol must be restricted to the same maximum weekly quantities as for the general population, that is, 14 units (women) and 21 units (men), with 1–2 alcohol-free days per week. In these quantities, alcohol has cardioprotective effects.

Fats

Since obesity is a major problem in type 2 diabetes and fats contain more than twice the energy content per unit mass than either carbohydrate or protein, consumption of fats should be limited. Monounsaturated fats have a lower atherogenic potential and are therefore recommended as the main source of dietary fat. Intake of fat should be less than 35% of total energy consumption, with saturated and trans-unsaturated fats accounting for less than 10% of energy intake and monounsaturated fats providing 10–20%.

Examples of monounsaturated fats are olive oil and rapeseed (also known as canola) oil. Saturated fats are chiefly of animal origin, for example, beef, pork, lamb, whole milk products with some found in plants, for example, cocoa butter, coconut oil and palm oil. Trans-unsaturated fats are found in hydrogenated vegetable oils and hard margarines. N-6 polyunsaturated fats such as cornflower, sunflower, safflower, soyabean oil and seed oils should account for less than 10% of energy intake, and n-3 polyunsaturated fats (fish oils) should be eaten as fish, rather than fish oil supplements, once or twice a week.

Protein

For adults without nephropathy, protein intake is recommended as less than 1 g/kg of body weight, equivalent to about 10–20% of total energy intake. For those with nephropathy, protein intake may need to be further restricted, but this requires expert dietetic advice and supervision.

Fibre

There is no quantitative dietary recommendation for fibre intake. Dietary fibre has useful properties in that it is physically bulky, and it delays the digestion and absorption of complex carbohydrates, thereby minimising hyperglycaemia. For the average person with type 2 diabetes, 15 g of soluble fibre from fruit, vegetables or pulses is likely to produce a 10% improvement in fasting blood glucose, glycated haemoglobin and low-density lipoprotein cholesterol (LDL-C). Insoluble fibre from cereals, wholemeal bread, rice and pasta has no direct effect on glycaemia or dyslipidaemia, but it has an overall benefit on gastro-intestinal health and may help in weight loss by promoting satiety.

Salt

Sodium chloride should be limited to a maximum of 6 g/day. A reduction in salt intake from 12 to 6 g/day has been shown to produce a reduction in systolic blood pressure of 5 mmHg and a reduction of 2–3 mmHg in diastolic pressure.

Obesity management in type 2 diabetes

Obesity management is a very important issue in type 2 diabetes owing to the insulin resistance which occurs as a consequence of excess adipose tissue. Whilst any loss of weight in those who are overweight or obese is of benefit in diabetes in that it is associated with an improvement in dyslipidaemia, hypertension and glycaemic control, bariatric surgery can lead to profound improvements. Recent studies have shown that laparoscopic banding can induce remission of type 2 diabetes in 48% of individuals and Roux-en-y bypass procedures can induce remission in 84%. More importantly, bariatric surgery is associated with a 92% relative risk reduction in diabetes-specific mortality and consequently should be offered to those with diabetes who have a BMI of 35 kg/m² or higher (National Institute for Health and Clinical Excellence, 2006). For those with a BMI of 28 kg/m² or greater, it is recommended that the pancreatic and gastric lipase inhibitor, orlistat, be considered as the modest weight reduction which can be achieved with this agent yields benefits in diabetes control.

Insulin therapy in type 1 diabetes

All patients with type 1 diabetes require treatment with insulin in order to survive. Exogenous insulin is used to mimic the normal physiological pattern of insulin secretion as closely as possible for each individual patient. However, a balance is required between tight glycaemic control and hypoglycaemia risk. If the risk of hypoglycaemia is high, then it may be necessary to aim for less tight glycaemic control. There is a wide variety of insulin preparations available which differ in species of origin, onset of action, time to peak effect and duration of action (Table 44.3).

Table 44.3 Insulin preparations

Species of origin

Until the 1980s, insulin was obtained and purified from the pancreas of pigs and cows. Human sequence insulins have subsequently been developed using recombinant DNA technology and are now the most common insulins in use. Many of the animal-derived products have been withdrawn, but some animal insulins continue to be available. Porcine insulin only differs from human insulin in one amino acid at the end of the B chain (position B30). Human insulin may be produced semi-synthetically by enzymatic modification of porcine insulin (emp). However, most human insulin is manufactured using genetic engineering and recombinant DNA technology. This is done by inserting either synthetic genes for the insulin A chain and B chain, or the proinsulin gene, or a proinsulin-like precursor into Escherichia coli (crb, prb) or yeast cells (pyr). The cells are fermented, resulting in large amounts of the recombinant protein, which is then converted into insulin and purified. More recently, human insulin analogues have been developed through genetic and protein engineering, to produce insulin molecules with differing pharmacokinetic properties.

It is now standard practice to commence all patients requiring insulin on human insulin. In those who have been changed from animal to human insulin, there has been concern that human insulin may be associated with an increased risk of hypoglycaemic unawareness, although current evidence suggests that this is unlikely if the switch is done appropriately. Human insulin may be more potent dose for dose than animal insulin due to the formation of antibodies. Conventionally, doses are reduced by 25% or more when changing from animal to either human or analogue insulin.

Insulin preparations

The onset of action, peak effect and duration of action are determined by the insulin type and by the physical and chemical form of the insulin.

Fast-acting insulins

Conventional fast-acting insulins are soluble insulins (also known as neutral insulins). After subcutaneous injection, soluble insulin appears in the circulation within 10 min. The concentration rises to a peak after about 2 h and then declines over a further 4–8 h. This absorption curve can be contrasted with the physiological insulin concentration curve, where peak concentrations are reached 30–40 min after a meal and decline rapidly to 10–20% of peak levels after about 2 h.

The fast-acting recombinant insulin analogues (insulin lispro, insulin aspart and insulin glulisine) are more rapidly absorbed than the non-analogue soluble insulins and have a shorter duration of action. The analogues therefore offer more flexibility. They are more convenient for some patients as they can be given immediately before a meal rather than the 30 min before recommended for human soluble insulin. Another benefit is a reduced risk of hypoglycaemia because of the shorter duration of action. These pharmacokinetic differences arise as the short-acting analogues remain as monomers (single units) unlike regular soluble human insulins which self-associate into a hexameric (6-unit) form. Hexamers need to dissociate into dimers and monomers to be readily absorbed from subcutaneous tissue, which causes delayed absorption.

Intermediate-acting insulins

Conventional intermediate-acting insulins are insoluble, cloudy suspensions of insulin complexed with either protamine (also known as isophane or NPH insulin) or zinc (lente insulin).

Over time, insulin dissociates from the protamine, which gives the preparation its extended activity. The onset of action is usually 1–2 h with the peak effect being seen at 4–8 h. There is considerable inter-patient variation in the duration of action, but it usually requires twice-daily administration to adequately cover a 24-h period. Protamine insulin and soluble insulin do not interact when mixed together. Therefore, ready-mixed (biphasic) preparations are available containing both isophane and soluble insulin.

Long-acting insulins

More recently, long-acting insulin analogues such as insulin glargine and insulin detemir have been developed using recombinant DNA technology. They both have a duration of action of about 24 h, a more predictable, flat profile of action with no pronounced peaks and less inter- and intra-subject dosing variability.

Insulin glargine differs from human insulin as two arginine molecules have been added to the B chain at the C-terminal end. This alters the isoelectric point from pH 5.4 to 6.7. Also, the neutral amino acid glycine replaces the asparagine residue at position A21. The changes mean that insulin glargine remains soluble at a slightly acidic pH. The product is buffered at a pH of 4. Once it is injected into subcutaneous tissue, it forms a microprecipitate in the more neutral surrounding pH. This allows slow absorption from the injection site.

Insulin detemir has a long duration of action and is formulated at neutral pH. It differs from human insulin by omission of the amino acid threonine at position B30 and the attachment of a fatty acid chain (myristic acid) to lysine at position B29. The modification allows the insulin molecule to reversibly bind to albumin, via the fatty acid chain, following absorption from subcutaneous injection. This reduces the amount of free, active insulin detemir (bound insulin is inactive). The long duration of action is produced by dissociation of the insulin molecule from albumin.

Insulin delivery

All the currently licensed insulin products are available only by injection. An inhaled insulin product available for a short period of time is no longer obtainable. The subcutaneous route is routinely used for maintenance therapy, as opposed to the intravenous route which is sometimes used in hospital. Insulin can be injected subcutaneously into the outer aspect of the thigh, abdominal wall, buttocks or upper arm. However, injecting into the arm is incredibly difficult and is therefore not usually a site of choice. The main advantages associated with subcutaneous injection are accessibility, which allows most patients to administer their own insulin. However, this route cannot be regarded as physiological as it delivers insulin to the systemic rather than portal circulation.

A small number of patients still use disposable plastic syringes with insulin from a vial as their means of insulin administration, although the vast majority now use pen injection devices. Insulin pens may either be refillable or disposable. Although not in themselves improving diabetic control, they are popular amongst users since they are compact and more convenient as they remove the need to draw up insulin from a vial.

Intravenous insulin delivery should be used in the management of ketoacidosis and hyperosmolar states. The intravenous route is also preferable for diabetic patients due to have major surgery and who may be ‘nil by mouth’ after surgery. The short half-life of insulin (3–5 min) means that changes in infusion rate have a rapid effect on insulin action and glycaemic control. Intravenous insulin is commonly delivered as either a continuous insulin infusion at a variable rate or an infusion made with a fixed amount of insulin, with glucose and potassium in the same bag. The former is also commonly referred to as a ‘sliding scale’ insulin regimen, in which the rate of infusion is adjusted according to frequent blood glucose readings, usually hourly. It is administered with a co-infusion of glucose (with potassium, unless patient is hyperkalaemic). The latter type of insulin infusion is known as GKI (glucose, potassium, insulin), GIK or the Alberti regimen. Intravenous insulin regimens are not routinely recommended for patients who are eating and drinking.

Insulin regimens

Standard insulin regimens for managing type 1 disease vary between two to five injections daily. They must be tailored to the individual patient and will depend on lifestyle, willingness to achieve the best control and ability to cope with both injecting insulin and subsequent monitoring of blood glucose. The chosen insulin regimen is negotiated between patient and health care professional and may change throughout life according to priorities and patient preference. Starting doses of insulin and the ratio of short- to intermediate-acting insulin are very variable. In patients who are very active, such as manual workers and those who exercise regularly, the starting dose should be kept low to reduce the risk of significant hypoglycaemia.

Mealtime plus basal regimens

The best control for type 1 diabetes may be attained using a mealtime plus basal regimen, also referred to as a basal-bolus regimen. This mimics normal physiological insulin release more closely than other regimens. A mealtime plus basal regimen requires mealtime injections of insulin with a fast-acting preparation, preferably with an analogue, plus one or two injections of a basal (intermediate- or long-acting) insulin. This may require up to five injections a day. As a general rule, with this regimen, the soluble insulin injections given before each meal usually comprise 40–60% of the total daily dosage. Some individuals may benefit from exogenous insulin delivery via a continuous subcutaneous insulin infusion administered via a pump worn on their person. The pump can be programmed to give a different basal rate of infusion at different times of day, and boluses are then provided by the pump at mealtimes. There are specific indications for pump therapy (National Institute for Health and Clinical Excellence, 2008).

These regimens offer the most flexibility of dosing and eating habits, and often better blood glucose control. A number of patients have been taught to count mealtime carbohydrates and calculate their own insulin dose on the basis of the preprandial blood glucose concentration, which allows greater scope for ‘normal eating’. An example is the DAFNE (dose adjusted for normal eating) programme (DAFNE Study Group, 2002), in which patients are required to attend structured, group education on five consecutive days.

The disadvantage of mealtime plus basal regimens is that they require multiple injections, unless a pump is in situ, and require regular blood glucose monitoring and the ability of the patient to match insulin doses according to carbohydrate intake, exercise levels and prevailing glucose levels. For some people, this is either too difficult or unsuitable.

Twice-daily regimens

The mealtime plus basal regimen may be too hard for some, for example, school-age children, to manage. In this type of situation, a twice-daily regimen may be more suitable. The simplest and most effective twice-daily regimens use premixed insulin, comprising a short- or rapid-acting plus an intermediate-acting insulin. Regular human insulin mixes and analogue mixes are available. The regular human insulin mixes should be given 30 min before breakfast and 30 min before the evening meal, whereas analogue mixes may be given immediately before these meals. The longer-acting component of the insulin mix given at breakfast time must span the lunchtime meal and the evening dose must bridge the night time. Twice-daily regimens using intermediate- or long-acting insulins alone are not sufficient for maintenance control of type 1 diabetes. Occasionally, they are used in newly diagnosed patients who are not acutely ill, adding the short-acting preparation if and when indicated by self-monitoring.

Adjusting the insulin dose

The information on which insulin dosage adjustment is based should be derived from blood glucose self-monitoring and the incidence and timing of hypoglycaemia. On twice-daily fast- and intermediate-acting insulin regimens, the soluble insulin may be considered as acting up to the next meal or to bedtime, while the extended-acting insulins act up to the next injection. The glucose concentration at the end of the period can be taken as a measure of the appropriateness of the relevant dose.

Adjustments to a dose of insulin should depend on the degree of insulin resistance present. In order to determine a suitable adjustment dose, the effect of other dosage adjustments in the same patient should be taken into consideration, as should the total insulin dose. For example, a 2-unit dose increase in someone taking 6 units of insulin would be a 33% dosage increase; however, a 2-unit dose increase in someone taking 60 units of insulin would be proportionately much less and make less impact.

Storage of insulin

Insulin formulations are stable if kept out of light, and they are not subject to freezing or extremes of heat. Loss of potency of 5–10% occurs in vials kept at high ambient room temperatures for 2–3 months. Insulin should therefore be stored in a domestic refrigerator except for the vial(s), cartridge(s) or pens in current use which, depending on the individual preparation, may be stable for 4–6 weeks (see manufacturers’ recommendations). When pen injector devices are in use, they should never be stored in a refrigerator as there have been reports of devices ‘seizing up’ when stored in the cold. Also, the injecting of cold or refrigerated insulin is undesirable because it is more painful and the insulin absorption profile is altered.

Adverse effects of insulin

Hypoglycaemia is a common physiological complication of insulin therapy and is often a source of great anxiety to patients and carers. The symptoms (see Box 44.2) may occur at different blood glucose levels in different individuals.

Thickening of subcutaneous tissues can occur at injection sites because of recurrent injection in the same area, known as lipohypertrophy. As well as looking unsightly, it can result in impaired and erratic insulin absorption, leading to poor glycaemic control. The solution is to rotate injection sites. Bruising is usually a sign of superficial injections. Localised skin reactions occasionally occur but resolve even with continued use of the same insulin preparation.

Systemic allergic reactions rarely occur with the current universal use of highly purified insulins. Though not usually species specific, it is worthwhile trying insulin of a different species if allergy occurs. Also, some patients may experience allergy to the excipients within the insulin product or to the needle used for administration. If this is suspected, it is helpful to seek the advice of an immunologist and try an insulin product with different excipients.

Management of type 2 diabetes

About 80% of patients with type 2 diabetes are overweight at diagnosis, and this is known to cause insulin resistance. This means that higher doses of medication may be required to control blood glucose levels. Advice on weight loss through increased physical exercise and calorie restriction, in addition to education about general healthy eating, is required. Targets for weight should be to maintain a normal BMI of between 20 and 25 kg/m² or a waist circumference of less than 88 cm in women and 102 cm in men, which lowers the risk of developing insulin resistance. In those who are already overweight or obese, however, an achievable target of 10–15% body weight loss should be discussed as, if achieved, this will have significant benefits in overall diabetes control.

Some people are able to normalise their glycaemic control by weight loss and attention to diet (diet controlled). Nevertheless, such individuals still invariably have diabetes and are at risk of developing diabetic complications. Hyperglycaemia may still occur, especially in times of stress or if dietary control is lost, and consequently they should be monitored regularly.

For over 75% of people with type 2 diabetes, dietary measures and exercise alone do not produce adequate glycaemic control and oral hypoglycaemic therapy is required. Within 3 years of diagnosis, a large majority of patients will require oral drug therapy. In the UK, there are six classes of oral agents currently available: a biguanide (metformin), sulphonylureas (glibenclamide, gliclazide, glimepiride, glipizide, tolbutamide), meglitinides (repaglinide and nateglinide), a thiazolidinedione (pioglitazone), an α-glucosidase inhibitor (acarbose) and the dipeptidyl peptidase-4 inhibitors (saxagliptin, sitagliptin and vildagliptin).

Acarbose has been poorly tolerated in trials, with only 39% of those receiving the drug still taking it after 3 years. The main reason for non-adherence appears to be flatulence. Acarbose is rarely prescribed in the UK but is popular in other countries such as Germany. Metformin remains the cornerstone of oral treatment for type 2 diabetes. The sulphonylureas and meglitinides are known as insulin secretagogues, since they both enhance secretion of insulin from the pancreatic β-cells. The relatively recent introduction of the dipeptidyl peptidase inhibitors (DPP-4 inhibitors) has been welcomed and is a useful new class of drug, particularly for those in whom weight is a problem, since they do not cause weight gain like many of the other drug treatments. Likewise, the incretin mimetics, another new class of injectable drugs, are often helpful for the obese population since they are associated with weight reduction.

In type 2 diabetes, the progressive decline in β-cell function with time and increasing insulin resistance means people with this disease show a progressive loss of glycaemic control and usually require two or three drugs to maintain control before ultimately requiring insulin.

The factors used to select a particular treatment include the patient’s clinical characteristics, such as their degree of hyperglycaemia, weight and renal function (Fig. 44.3). In acutely ill people with significant hyperglycaemia, insulin therapy may well be required, albeit transiently because acute illness leads to an increase in stress hormones, all of which are anti-insulin.

Fig. 44.3 Algorithm for the treatment of glycaemic control in type 2 diabetes. (1) or individually agreed target, (2) with active dose titration (3) see the NICE clinical guideline on obesity (www.nice.org.uk/CG43), (4) offer once-daily sulphonylurea if adherence is a problem, (5) only continue DPP-4 inhibitor or thiazolidinedione if reduction in HbA_1c of at least 0.5 percentage points in 6 months, (6) only continue exenatide if reduction in HbA_1c of at least 1 percentage point and weight loss of at least 3% of initial body weight in 6 months, (7) with adjustment for other ethnic groups, (8) continue with metformin and sulfylurea (and acarbose if used), but only continue other drugs that are licensed for use with insulin. Review the use of sulphonylurea if hypoglycaemia occurs, (9) DPP-4 inhibitor refers to saxagliptin, sitagliptin or vidagliptin, (10) thiazolidinedione refers to ploglitazone.

Biguanides

Metformin is the only biguanide available in the UK. The mechanism of action of biguanides is still not completely understood. However, the principal mode of action is via potentiation of insulin action at an unknown intracellular locus, resulting in decreased hepatic glucose production by both gluconeogenesis and glycogenolysis. Metformin also stimulates tissue uptake of glucose, particularly in muscle, and is thought to reduce gastro-intestinal absorption of carbohydrate. The action of metformin does not involve stimulation of pancreatic insulin secretion and therefore it is still a beneficial agent when β-cell function has declined. Another advantage of metformin over insulin secretagogues, and sulphonylureas, in particular, is that it does not cause hypoglycaemia and is not associated with weight gain. Metformin has a short duration of action, with a half-life of between 1.3 and 4.5 h, and does not bind to serum proteins. It is not metabolised and is totally renally eliminated.

It has been shown that diabetes-related death was reduced by 42% in overweight subjects who took metformin for 10 years, compared to those who took conventional therapies such as a sulphonylurea or insulin. Myocardial infarction was also reduced by 39% over the 10-year follow-up period. Consequently, metformin has become the first-line therapy for glycaemic control when oral agents are indicated especially in overweight and obese patients.

Adverse effects

The most common adverse effects of metformin, affecting about a third of patients, result from gastro-intestinal disturbances including anorexia, nausea, abdominal discomfort and diarrhoea. In some patients, diarrhoea can be extreme and can preclude metformin use. However, the gastro-intestinal side effects are usually transient and can be minimised by starting with a low dose, increasing the dose slowly and administering the drug with or after food. A suggested regimen is to start with 500 mg daily for 1 week, then 500 mg twice daily for 1 week, increasing the dosage at weekly intervals until the desired glycaemic response is achieved or intolerance occurs. The maximum licensed dose is 3 g/day, but doses of more than 2 g/day often cause intolerance. If the initial dose of 500 mg daily causes side effects, then some prescribers reduce the starting dose to 250 mg daily for a week. This may be difficult for some patients as the 500 mg strength tablets are not scored and are not easy to halve.

Two modified-release metformin preparations are now available and permit once-daily dosing. Clinical evidence suggests that these formulations cause fewer problems with gastro-intestinal side effects. The maximum licensed dose of the modified-release preparations (2 g daily) differs from the standard preparation.

The two previously available biguanides, phenformin and buformin, were withdrawn due to deaths associated with lactic acidosis. Lactic acidosis is a rare complication with metformin, with an estimated incidence of five cases per 100,000 patient-years. However, it is potentially life threatening. Patients at most risk are those with renal insufficiency in whom the drug accumulates, individuals with co-existing conditions where lactate accumulates, and those who are unable to metabolise lactate. In practice, metformin should not be prescribed for patients who have renal impairment (eGFR <45 mL/min/1.73 m²) and should be stopped in anyone with an eGFR <30 mL/min/1.73 m². Metformin should also be stopped in severe liver disease, uncontrolled cardiac failure or severe pulmonary insufficiency. It should be withdrawn in patients with severe intercurrent illness, for example, acute myocardial infarction or septicaemia, or those undergoing major surgery or requiring investigation using radiographic contrast media and should only be restarted once renal function has been evaluated and determined as within acceptable limits.

Role of metformin

Metformin is useful in obese patients with diabetes as it does not cause weight gain. If there are no contraindications, it can be used in conjunction with diet as second-line therapy in patients not adequately controlled on diet alone. As it has a different mode of action to the sulphonylureas, meglitinides, thiazolidinediones, α-glucosidase inhibitors and DPP-4 inhibitors, it can be valuable when prescribed in combination.

Sulphonylureas

Mode of action

The major action of this class of drug relies on the ability of the pancreas to secrete insulin and hence requires functioning β-cells to exert a beneficial effect. Sulphonylureas lower blood sugar by increasing pancreatic β-cell sensitivity to glucose, allowing more insulin to be released from storage granules for a given glucose load. Sulphonylurea therapy is also associated with increased tissue sensitivity to insulin, resulting in improved insulin action. Studies also suggest that sulphonylureas may promote an increased systemic bioavailability of insulin due to reduced hepatic extraction of the insulin secreted from the pancreas.

Pharmacokinetics

The pharmacokinetic parameters of oral hypoglycaemic agents are shown in Table 44.4. Chlorpropamide is the slowest and longest acting agent, but it is now very rarely used. Although glibenclamide has been shown to have a short elimination half-life, it has a prolonged biological effect, which may be explained by slower distribution and the existence of a deep compartment, possibly the islet cells. All sulphonylureas are metabolised by the liver to some degree and some may have active metabolites.

Table 44.4 Pharmacokinetic properties of oral hypoglycaemic agents

Choice of drug

There are many factors which influence the choice of sulphonylurea. These may relate to the drug itself, the patient or the prescriber. There are few well-controlled long-term clinical comparisons between sulphonylureas. It would appear that, when dosage is individualised and governed by the effect on fasting blood glucose, there is little or no difference in clinical efficacy between different agents. In general, if a patient is not well controlled on the maximum dosage of one sulphonylurea, it is not worthwhile changing to another one.

Adverse effects

The frequency of adverse effects from sulphonylureas is low. They are usually mild and reversible on drug withdrawal (Table 44.5). The most common adverse effect is hypoglycaemia, which may be profound and long lasting. Hypoglycaemia due to sulphonylureas is often misdiagnosed, particularly in the elderly. The major risk factors for the development of hypoglycaemia include use of a long-acting agent, increasing age, renal or hepatic dysfunction and inadequate carbohydrate intake. The major side effect is, however, weight gain.

Table 44.5 Adverse effects of sulphonylureas

Adverse effect	Comments
Gastro-intestinal	Affects approximately 2%
	Most commonly nausea and vomiting
	Dose related
	Advise patient to take with or after food
Dermatological	Affects 1–3%
	Usually occur within the first 2–6 weeks
	Most commonly: generalised photosensitivity, pruritus, maculopapular rash
	May require discontinuation of drug
	Cross-sensitivity between sulphonylureas is common
	Rare cases of severe allergic reactions, for example, erythema multiforme Stevens–Johnson syndrome
Haematological	Rare cases of fatal agranulocytosis or pancytopenia
	Other haematological effects usually reversible on discontinuing drug
	Some reports of reversible haemolytic anaemia
Hepatic	Mild, reversible elevation of liver function tests
	Cholestatic jaundice
	Usually a hypersensitivity reaction associated with fever, rash and eosinophilia
Cardiovascular	Possible excess of cardiovascular mortality in patients treated with tolbutamide (not proven)
Hypothyroidism	Association not proven
Hypothyroidism	May be rare cases
Alcohol flush	Rarely seen with sulphonylureas other than chlorpropamide
Alcohol flush	Change to another agent
Syndrome of inappropriate antidiuretic hormone (SIADH)	Chlorpropamide and, to a lesser extent, tolbutamide enhance the effect of ADH on the kidney
	Results in hyponatraemia
	Risk factors are increasing age, congestive cardiac failure and diuretic therapy
Hypoglycaemia	The most common adverse effect and may be severe and prolonged
	Highest incidence with chlorpropamide and glibenclamide
	All sulphonylureas and meglitinides have been implicated
	Risk factors include increasing age, impaired renal or hepatic function, reduced food intake, weight loss
	Decrease dose, change to a shorter-acting agent or discontinue sulphonylurea therapy

Other adverse effects are rare; blood dyscrasias occur in 0.1% of patients and rashes in up to 3%. Chlorpropamide can produce troublesome flushing after ingestion of alcohol, and about 5% of patients develop hyponatraemia due to its effect on increasing renal sensitivity to antidiuretic hormone (ADH). Most patients are asymptomatic with this problem, but occasionally severe hyponatraemia is observed. However, as mentioned earlier, this agent is no longer used in routine practice.

Sulphonylurea dosage

The dosage should be individualised for each patient. The lowest possible dose required to attain the desired levels of blood glucose, without producing hypoglycaemia, should be used. Treatment should start with a low dose and be increased if necessary approximately every 2 weeks. For many agents, the maximum effect is seen if the dose is taken half an hour before a meal, rather than with or after food. The number of daily doses required will depend on the agent used and the total daily dose. For several drugs, it becomes necessary to administer the drug two or three times daily when the dose is increased.

A modified-release preparation of gliclazide is available. A dose of 30 mg of the modified-release product is equivalent to 80 mg of the standard-release preparation.

Drug interactions

Several drugs can interfere with the efficacy of sulphonylureas by influencing either their pharmacokinetics or pharmacodynamics, or both. Despite much literature about displacement interactions with sulphonylureas, the clinical significance is doubtful. Many reported cases involve first-generation agents, which have a different protein-binding site from the second-generation agents, which bind in a non-ionic fashion and are not readily displaced. Ingestion of alcohol can cause hypoglycaemia in itself and can also prolong the hypoglycaemic effect of sulphonylureas.

Meglitinides

The meglitinides are insulin-releasing agents (insulin secretagogues), also called ‘post-prandial glucose regulators’. They are characterised by a more rapid onset and shorter duration of action than sulphonylureas. Their site of action is pharmacologically distinct from that of the sulphonylureas. Repaglinide, a benzoic acid derivative, was the first member of the class. It is licensed for use as a single agent when diet control, weight reduction and exercise have failed to regulate glucose levels, or in combination with metformin. Nateglinide was introduced later and is a derivative of the amino acid d-phenylalanine. Nateglinide is only licensed for combination therapy with metformin when metformin alone is inadequate.

Mode of action

Like the sulphonylureas, the meglitinides stimulate first-phase insulin secretion by inhibiting ATP-sensitive potassium channels in the membrane of the pancreatic β-cells. This causes depolarisation and gating of the calcium channels (which are voltage sensitive), increasing the intracellular concentration of calcium and stimulating insulin release. The release of insulin only occurs in the presence of glucose. As glucose levels drop, less insulin is secreted. Conversely, if carbohydrates are consumed and glucose levels rise, insulin secretion is enhanced.

Pharmacokinetics

The pharmacokinetic properties of the meglitinides confer a rapid onset and a short duration of action. The individual parameters are shown in Table 44.4. The meglitinides are extensively metabolised in the liver, repaglinide by oxidative biotransformation and direct conjugation with glucuronic acid. The cytochrome P450 enzymes CYP2C8 and CYP3A4 have been shown in vitro to be involved its metabolism. Nateglinide is metabolised predominantly by cytochrome P450 enzyme CYP2C9 and to a lesser extent by CYP3A4. Repaglinide has no active metabolites, but nateglinide has partially active metabolites, one-third to one-sixth the potency of the parent compound. The meglitinides should be taken immediately before main meals, although the time can vary up to 30 min before a meal. The pharmacokinetic profile of meglitinides offers some advantages in patients with poor renal function or irregular eating habits.

Adverse effects

The meglitinides may cause a range of side effects, most commonly hypoglycaemia, visual disturbances, abdominal pain, diarrhoea, constipation, nausea and vomiting. More rarely, hypersensitivity reactions can occur as well as elevation of liver enzymes. The meglitinides may also cause weight gain.

Dosage

Drug interactions

Drugs which induce or inhibit the cytochrome P450 enzymes CYP2C8 and CYP3A4 interact with repaglinide. Examples of drugs which enhance or prolong the hypoglycaemic effect include gemfibrozil, clarithromycin, ketoconazole, itraconazole, trimethoprim, other hypoglycaemic drugs, monoamine oxidase inhibitors, non-selective β-blockers, ACE inhibitors, salicylates, NSAIDs, octreotide, alcohol and anabolic steroids. Drugs which induce cytochrome P450 enzymes may also interact, for example, rifampicin and phenytoin, and may decrease repaglinide serum levels. Drugs that inhibit CYP2C9 may interact with nateglinide. Drugs that may enhance or prolong the hypoglycaemic effect include ACE inhibitors, gemfibrozil and fluconazole. The hypoglycaemic effects of nateglinide may be reduced by diuretics, corticosteroids and β-blockers.

Role of meglitinides

The meglitinides are an effective hypoglycaemic therapy in type 2 diabetes. They may be most beneficial in patients who experience problems with post-prandial glucose elevation and as single therapy in patients who eat at unpredictable times or have a tendency to miss meals. However, no outcome studies have been undertaken, and so their long-term use has not been proven to be more effective than the less expensive sulphonylureas. Therefore, their exact place in therapy is currently unclear.

Thiazolidinediones

Research into the action of the thiazolidinediones, also known as glitazones, has led to greater understanding of the development of type 2 diabetes. Only one glitazone, pioglitazone, is currently available following the removal of rosiglitazone from the UK market in 2010. Pioglitazone has been shown to have a significant benefit on macrovascular morbidity and mortality, demonstrating the benefit of a glucose-lowering agent on macrovascular disease (Dormandy et al., 2005).

Mode of action

The glitazones act as agonists of the nuclear peroxisome proliferator-activated receptor-γ (PPAR-γ). PPAR-γ is mostly expressed in adipose tissue but is also found in pancreatic β-cells, vascular endothelium and macrophages. It is also expressed weakly in those tissues that express predominantly PPAR-α, for example, skeletal muscle, liver and heart. The thiazolidinediones lower fasting and post-prandial glucose levels in addition to lowering free fatty acid and insulin concentrations. They enhance insulin sensitivity and promote glucose uptake and utilisation in peripheral tissues. They also suppress gluconeogenesis in the liver and, by increasing insulin sensitivity in adipose tissue, suppress free fatty acid concentrations. In addition, patients with IGT have shown increased insulin secretory responses. However, they do not have a direct effect on insulin secretion. The indirect effects of glitazones on adipose tissue are due to alterations in the regulation of gene expression. Various adipokines (adiponectin, tumour necrosis factor-α, resistin and 11β-hydroxysteroid dehydrogenase 1) are regulated by PPAR-γ agonists in animal studies. Other effects of glitazones on the vasculature include antiatherogenic effects thought to be caused by a reduction in the inflammatory response, decrease in vasoconstriction and an increase in plaque stability.

Pharmacokinetics

Pioglitazone is metabolised extensively in the liver to both active and inactive metabolites (see Table 44.4).

Adverse effects

Pioglitazone has been associated with weight gain and oedema, particularly in patients with hypertension and congestive cardiac failure. Since the thiazolidinediones can lead to a worsening of heart failure that may be fatal, pioglitazone should not be used in patients with a previous history of or pre-existing heart failure. Combination therapy with insulin and thiazolidinediones has been found to result in a higher incidence of oedema. There is also an increased risk of bone fracture with pioglitazone, and it should be used with caution in post-menopausal women. Anaemia occurs in about 1% of patients and is seen as a small decrease in the haemoglobin concentration during the first 4–12 weeks of therapy. However, it is suggested this decrease is due to dilutional effects caused by an increase in serum volume. Pioglitazone caused weight gain of around 3.5 kg during clinical trials. Some patients also experience headache, abdominal pain, myalgia and upper respiratory infection. Pioglitazone may also elevate liver transaminases.

Rosiglitazone was withdrawn from the UK market because use was associated with an increased risk of cardiovascular disorders, including heart attacks and heart failure. Another thiazolidinedione, troglitazone, was withdrawn from the UK market in 1997 because of liver failure linked to a toxic metabolite of troglitazone.

Dosage

Pioglitazone is started at a dose of 15 mg or 30 mg and may be increased to 45 mg once daily. In combination with metformin or a sulphonylurea, the current dose can be continued. Administration of pioglitazone may be either with or without food. Dosage adjustment is not necessary in patients with mild or moderate renal impairment or in the elderly. However, neither agent should be used in patients with severe renal impairment or in those with hepatic impairment.

Drug interactions

Pioglitazone is metabolised by cytochrome P450 CYP3A4. Therefore, drugs which induce or inhibit this enzyme interact with pioglitazone. Ketoconazole, itraconazole, erythromycin and fluconazole increase serum concentrations, while rifampicin and phenytoin decrease serum levels of pioglitazone.

Role of thiazolidinediones

Thiazolidinediones improve glycaemic control in patients, especially in those with insulin resistance, by reducing HbA_1c levels up to 1.5% compared to sulphonylurea or metformin alone.

Glitazones should be used as third-line therapy after life style modification and the use of metformin or a sulphonylurea as monotherapy. However, if glycaemic control remains poor, pioglitazone can be used either with metformin if treatment with a sulphonylurea is unsuitable, with a sulphonylurea if treatment with metformin is unsuitable, or with metformin and a sulphonylurea if insulin is unsuitable (National Institute for Health and Clinical Excellence, 2009). Treatment should only be continued if, after 6 months of treatment, the HbA_1c has reduced by 0.5% of its starting value.

Monotherapy with a thiazolidinedione may be a valuable treatment option for patients who are known to be insulin resistant. Triple therapy can be an alternative to transferring a patient to insulin, but the modest reduction in HbA_1c usually means that many patients will eventually require insulin. It is important to be aware that owing to their mode of action involving changes in gene transcription, thiazolidinediones take up to 3 months to have their maximum effect on glycaemic control.

α-glucosidase inhibitors

Acarbose reduces carbohydrate digestion by interfering with gastro-intestinal glucosidase activity. Although overall carbohydrate absorption is not significantly altered, the post-prandial hyperglycaemic peaks are markedly reduced. Acarbose is minimally absorbed in unchanged form from the gastro-intestinal tract.

Adverse effects

The most common adverse effect of acarbose is abdominal discomfort associated with flatulence and diarrhoea. These symptoms usually improve with continued treatment but can be minimised by starting with a low dose and titrating slowly.

Systemic adverse effects are rare but high doses have been associated with idiosyncratic elevations of serum hepatic transaminase levels. Patients titrated to the maximum dose of 200 mg three times daily should be closely monitored, preferably at monthly intervals for the first 6 months. If elevated transaminase levels are observed, reduction in dose or withdrawal of therapy should be considered.

DPP-4 inhibitors block the normal inactivation of incretins (glucagon like peptide-1 [GLP-1] and glucose-dependent insulinotropic peptide [GIP]). Incretins play a role in increasing endogenous insulin in response to a high glucose load, that is, post-prandially. They also reduce the amount of glucose produced by the liver when glucose levels are sufficiently high. By blocking DPP-4, these drugs prolong incretin activity and inhibit glucagon release. This in turn causes a decrease in blood glucose and an increase in insulin secretion.

Pharmacokinetics

DPP-4 inhibitors are predominantly renally excreted. However, there is also a degree of hepatic metabolism involved in the elimination process, which varies with each drug. Sitagliptin is mainly excreted as unchanged drug in the urine, with a small metabolic contribution from the liver via the cytochrome P450 system. The kidney is thought to be mainly responsible for metabolic hydrolysis of vildagliptin to an inactive compound. Although saxagliptin is mainly eliminated renally, some hepatic biotransformation does occur via the cytochrome P450 system (CYP3A4/5), which results in a metabolite with half the potency of the parent compound. The pharmacokinetic parameters of the DPP-4 inhibitors are detailed in Table 44.4.

Adverse effects

All three DPP-4 inhibitors have been linked to gastro-intestinal side effects and upper respiratory tract infection. DPP-4 inhibitors do not cause hypoglycaemia, but they have the potential to cause hypoglycaemia when prescribed with other agents that can produce this effect. Vildagliptin has been associated with rare reports of liver dysfunction. Therefore, it is recommended liver function tests are performed prior to starting treatment, at 3-monthly intervals for the first year and then periodically thereafter.

Drug interactions

DPP-4 inhibitors have a low potential for interaction with other medicines. However, as both sitagliptin and saxagliptin are metabolised by cytochrome P450 3A4/5 (and CYP2C8 for sitagliptin), they have the potential to interact with potent inducers or inhibitors of these enzyme. In practice, few drugs have been formally assessed. Therefore, general advice is to monitor blood glucose levels carefully if one of these drugs is co-prescribed with sitagliptin or saxagliptin.

Role of dipeptidyl peptidase-4 inhibitors

DPP-4 inhibitors are licensed for use as dual therapy with metformin, a sulphonylurea or a thiazolidinedione. Sitagliptin is also licensed as both mono- and triple therapy with metformin and a sulphonlyurea or a thiazolidinedione. Similar to the thiazolidinediones, it is recommended that a DPP-4 inhibitor is used as third-line therapy typically in those who still do not have adequate control or cannot tolerate treatment with metformin and/or a sulphonylurea. DPP-4 inhibitors are also useful if further weight gain is likely. However, if insulin resistance is a key factor, then treating with a thiazolidinedione may be a better choice.

Incretin mimetics

The incretin mimetics, as the name suggests, mimic the effects of incretins. The two currently licensed in the UK, exenatide and liraglutide, are only available as subcutaneously injectable products. Incretin mimetics have both been demonstrated to cause weight loss, which is a particularly beneficial effect in many patients with type 2 diabetes.

Mode of action

The incretin mimetics bind to and activate the glucagon-like peptide-1 (GLP-1) receptor, hence increasing insulin secretion, suppressing glucagon secretion, increasing satiety and slowing gastric emptying. All of these effects help to lower blood glucose levels.

Pharmacokinetics

Incretin mimetics have a longer duration of action than endogenous GLP-1. Exenatide is eliminated primarily by renal clearance. However, the specific organ responsible for liraglutide elimination has not been identified. Table 44.4 details additional pharmacokinetic parameters of the incretin mimetics.

Adverse effects

Both incretin mimetics have been associated with nausea, and other gastro-intestinal disturbances. However, once therapy has been established, the incidence of these side effects decreases. Acute pancreatitis has also been rarely reported. For this reason, both patients and their carers should be advised of the signs and symptoms of this complication and advised to stop taking the drug and seek medical attention immediately if they occur. Since the incretin mimetics cause glucose-dependent insulin release, hypoglycaemia is uncommon and in most cases can be attributed to other agents taken concurrently

Drug interactions

The potential for drug interactions with the incretin mimetics is low. However, as they can cause a delay in gastric emptying, they may influence the absorption of other medicines administered at the same time. For this reason, it is advised that any narrow therapeutic index drugs taken concomitantly are monitored carefully. Also, if other oral medicines are being taken where the threshold concentration is important (e.g. antibiotics), it is advised that these should be taken at least 1 h before exenatide. This may be assumed for liraglutide, although this is not specifically stated in the summary of product characteristics.

Role of incretin mimetics

Insulin therapy in type 2 diabetes

The younger age of onset of type 2 diabetes and tighter glycaemic targets mean that the majority of patients with type 2 diabetes progress to insulin therapy, since recent evidence confirms that long-term glycaemic improvement reduces the risk of both microvascular (Holman et al., 2008) and macrovascular (Turnbull et al., 2009) complications.

It is currently common practice to introduce insulin to an oral medication schedule, although if hypoglycaemia becomes a problem, then the oral medications may be reduced or stopped. A number of different insulin regimens for use in patients with type 2 diabetes are available, the most common of which include once-daily basal insulin, twice-daily biphasic (pre-mixed) insulin, or prandial insulin, using a rapid/short-acting insulin with meals. Until recently, there have not been any trial data to determine which insulin regimen is most effective in controlling blood glucose levels and minimising hypoglycaemia in patients with type 2 diabetes. However, recent work suggests that patients who have basal insulin or prandial insulin added to their oral therapy have better HbA_1c control than those who receive biphasic insulin. In addition, the basal insulin regimen is associated with fewer hypoglycaemic episodes and less weight gain than the other two regimens (Holman et al., 2009). Basal insulin should be titrated to achieve normal fasting glucose levels, and the patient may be taught this self-titration protocol (Davies et al., 2005).

In a lean patient (BMI <25 kg/m²), significant insulin deficiency is more likely and therefore from the outset of insulin treatment either a basal-bolus or twice-daily regimen may be preferred.

In type 2 patients who require temporary insulin during intercurrent illness, a soluble preparation such as Humulin S or Human Actrapid can be given two or three times daily with a small dose of isophane insulin at bedtime to control blood glucose quickly and eliminate symptoms. The dose is selected initially according to the patient’s previous insulin requirements, if any, and adjusted according to four times daily blood glucose measurements.

Treating hypertension

The co-existence of hypertension and diabetes dramatically increases the risk of microvascular and macrovascular complications. Most important is the increased risk of cardiovascular disease. Tight control of blood pressure may be a more effective method of preventing complications in patients with type 2 diabetes than tight glycaemic control. It is recommended that, for patients with type 2 diabetes, the target blood pressure should be <140/80 mmHg, or for those with pre-existing kidney, eye or cerebrovascular damage the target should be reduced to <130/80 (National Institute for Health and Clinical Excellence, 2009). First-line blood-pressure-lowering therapy should be a once-daily, generically prescribed angiotensin-converting enzyme (ACE) inhibitor. Exceptions to this are people of African-Caribbean descent, in whom first-line therapy should be an ACE inhibitor plus either a diuretic or a generic calcium-channel blocker.

In young patients with type 1 diabetes, the blood pressure target is <135/80 mmHg. However, if there is abnormal albumin excretion (renal disease) or two features of the metabolic syndrome are present, then the blood pressure target should be lowered to 130/80 mmHg with maximal doses of either an ACE inhibitor or ARB as first-line therapy.

Treating obesity

Obesity is a significant risk factor in the development of type 2 diabetes. Evidence suggests that for each kilogram increase in body weight, the risk of diabetes increases by 4.5%. It is estimated that almost one-fifth of the population has a BMI of over 30 kg/m² and is therefore obese.

Treatment of obesity may require modification of lifestyle (diet and exercise regimens) to reduce calorie intake and increase calorie utilisation, pharmacological intervention and/or surgery. Currently orlistat, a lipase inhibitor, is the only drug available on prescription for the treating of obesity. Orlistat can also be purchased in community pharmacies as a branded over-the-counter product. Both sibutramine and rimonabant were withdrawn because of unwanted side effects.

Orlistat is licensed for use with a ‘mildly hypocalorific diet’ to treat obese people (BMI >30 kg/m²) or overweight patients with a BMI >28 kg/m² and associated risk factors. It should be discontinued after 12 weeks if a 5% weight reduction since the start of treatment has not been achieved. Orlistat increases the amount of faecal fat excretion but is associated with a number of gastro-intestinal side effects such as oily leakage from the rectum, flatulence, faecal urgency and incontinence. A greater reduction in the incidence of type 2 diabetes has been observed in patients treated with both orlistat and lifestyle modification (Torgerson et al., 2004).

Over recent years, the use of bariatric surgery has increased in popularity, which is another treatment option for those who are seriously struggling with their weight with a BMI over 35–40 kg/m², who have other conditions caused by being overweight such as diabetes.

Patient care

Patient education

Patient involvement is paramount for the successful care of diabetes. This is highlighted in the national service standards for diabetes (Department of Health, 2002) which state that all patients, and carers, where appropriate, will be encouraged to develop a partnership with their clinicians to enable them to manage their diabetes and maintain a healthy lifestyle, often through shared care plans. Structured education for patients with type 2 diabetes is important and should be offered to every patient and/or their carer around the time of diagnosis. It is considered to be an integral part of diabetes care (National Institute for Health and Clinical Excellence, 2009).

Education will depend upon the individual patient and the availability of local resources. Individual tuition is preferable in the early stages after diagnosis and is usually delivered by a diabetes specialist nurse. The educational aspect of care is a gradual and ongoing process. At a later stage, group education can be effective and many patients appreciate and find support in meeting others who have the same disease. Many such programmes are multidisciplinary and involve doctors, nurses, dieticians, pharmacists and chiropodists. It is essential to involve the patient’s family and carers in the educational process. Patients can also obtain support and information from specialist organisations such as Diabetes UK (available at www.diabetes.org.uk/home.htm).

Patients require education and information about many subjects, ranging from general lifestyle advice through to knowledge about the medicines they are prescribed (Box 44.4).

Advice about the use of over-the-counter medications, diet, foot care products and diabetic food products is frequently requested.

Annual review

People with diabetes should attend their hospital clinic or primary care practice (if this service is offered locally) for an annual review to screen for diabetic complications. Monitoring and optimising glycaemic control are also undertaken although this should be done on a more regular basis, as should review of patients with known complications. The annual review is increasingly taking place in primary care, with referral to secondary care if required. The typical assessments undertaken at an annual review are described in Box 44.5.

Glycaemic management targets

The theoretical ideal for all patients with diabetes is to achieve normoglycaemia. As this is not always possible, the aim is to achieve the best possible control compatible with an acceptable lifestyle for the patient. In some patients, this may mean only symptomatic control, in others this may be tight control. In making this decision, the following factors should be considered: the patient’s age, motivation, intelligence, understanding, likely adherence, co-existing diseases, ability to recognise hypoglycaemia, duration of their diabetes and the presence/absence/severity of diabetic complications.

Targets for pre-meal blood glucose of between 4 and 7 mmol/L and post-meal values of <10 mmol/L may be set for most patients, provided there is no significant hypoglycaemia risk. An optimal HbA_1c target of 6.5% for most of those with type 2 diabetes has been suggested (National Institute for Health and Clinical Excellence, 2009). However, it is recommended that targets should be individualised and that for some a higher target would be more appropriate especially if there is a risk of hypoglycaemia at the lower target. For those with type 1 diabetes, the recommended target is less than 7.5% (National Institute for Health and Clinical Excellence, 2004).

The diabetes treatment goals in older people may be different and more conservative than in younger adults. For example, some elderly patients may have poor vision and limited manual dexterity, which may or may not be linked to a degree of cognitive impairment. Others may have multiple pathology and take a number of other medications. Therefore, the goals of therapy need to be both individual and realistic. In some people, they will involve only the optimisation of body weight, control of symptoms and avoidance of hypoglycaemia which has an increased risk of severe brain damage and may occur without the usual warning signs in the elderly. In others, reasonably tight control may be appropriate. There is, therefore, a difficult balance between the use of aggressive treatment with its associated risk of hypoglycaemia and the benefits of reducing complications to maintain an acceptable quality of life. Box 44.6 sets out some of the common therapeutic problems in diabetes.

Box 44.6 Common therapeutic problems in diabetes

• Achieving normoglycaemia or HbA_lc targets

• Achieving an acceptable balance between improving glycaemic control and minimising episodes of hypoglycaemia

• Achieving adequate control in type 2 diabetes with diet alone, especially in patients who are overweight

• Achieving blood pressure targets in patients with co-existing hypertension

• Ensuring adherence as drug regimens become complex, requiring multiple drug therapies

• Making insulin therapy acceptable to patients with type 2 diabetes, inadequately controlled with oral therapies

Monitoring glycaemic control

Clinic monitoring

There are several ways in which glycaemic control can be monitored in primary and secondary care. Estimates of average control are often useful. Glycation of minor haemoglobin components occurs in the blood, with the extent depending on both the amount of glucose present and the duration of exposure of the haemoglobin to glucose. Estimates of glycated haemoglobin (HbA_1c) provide an index of average diabetes control over the preceding 2–3 months, that is, the lifespan of a red blood cell. HbA_1c can be measured at any time, the patient does not need to be fasted, and levels are not normally affected by acute changes in therapy, diet or exercise. However, they may be lower in those with reduced red cell lifespan, for example, pregnancy or advanced renal failure. Serum fructosamine represents the glycation of all serum proteins and gives information about control over the preceding 3 weeks. As albumin is the major serum protein, hypoalbuminaemia may affect fructosamine levels. However, HbA_1c is the preferred marker for average glycaemic control.

Home monitoring

Type 1 diabetes

All patients treated with insulin should be offered home blood glucose monitoring (HBGM). Capillary blood is applied to a reagent strip which has been impregnated with enzymes, for example, glucose oxidase. Some home blood glucose monitoring methods require visual comparison of colour changes which correspond to various blood glucose concentrations. More commonly, meters may be used to give readings and are more convenient for people who are colour blind or who have poor eyesight. Home blood glucose monitoring enables patients and carers to make a direct assessment of the effect of changes in medicines, dietary habits, exercise and patterns of illness. It has the additional benefit that it can detect hypoglycaemia and, unlike urine testing in which glycosuria may only be detected some time after changes in blood glucose have occurred, it enables more accurate calculations of insulin doses.

Patients with type 1, who are by definition ketosis prone, should also know how and when to test their urine for ketones. This test need not be carried out as part of routine monitoring but is essential at times of intercurrent illness, especially when the blood glucose is ≥17 mmol/L.

Type 2 diabetes

Many people with type 2 diabetes are treated with diet alone or with oral hypoglycaemic agents, and unless they have problems with hypoglycaemia, urine glucose monitoring should be adequate. This is a simple non-invasive test that can detect hyperglycaemia but not hypoglycaemia. Home blood glucose monitoring is used by some patients with type 2, particularly if control is poor, if they are undergoing medication dose titration, if they are being treated with insulin or if they are prone to hypoglycaemia. It is also recommended that blood glucose monitoring be undertaken before driving and at times of intercurrent illness, when blood glucose levels may be particularly erratic.

Regardless of whether individuals with type 1 or type 2 diabetes are using home blood glucose monitoring or urine testing, it is important they are educated about what to do with the results; otherwise, there is little point in testing.

Case studies

Case 44.1

Mrs TM is a 36-year-old married lady who has type 1 diabetes. She undertook a home pregnancy test because she was feeling particularly nauseated in the mornings and her period was late. The test was positive confirming that she was pregnant. However, at 8 weeks, she experienced vaginal bleeding and abdominal pain. She attended the Accident & Emergency department, where a miscarriage was confirmed. Upon questioning, it was discovered that she had been taking folic acid 400 μcg daily for the previous 6 months but had not received any pre-conception diabetes care. Her most recent HbA_1c was 7.3% (56 mmol/mol). Her regular medications are ramipril 10 mg daily, simvastatin 40 mg daily, insulin glargine at night and insulin aspart three times daily with meals.

Questions

1. Why should women of childbearing age be offered advice about pregnancy?

2. What blood glucose targets should Mrs TM have been advised to aim for before and after conceiving?

3. Was she taking appropriate dietary supplements prior to conception?

4. What advice should she be given with respect to her regular medication?

Answers

1. Mrs TM should have been offered preconception advice prior to becoming pregnant because glucose control needs to be optimal to reduce the risks of miscarriage, congenital malformation, stillbirth and neonatal death associated with diabetes in early pregnancy. Preconception advice should also include information for the patient on how diabetes affects pregnancy and how pregnancy affects diabetes, what dietary supplements to take and advice on diabetes-related medicines that are unsafe to take during pregnancy.

2. Mrs TM should aim for an HbA_1c target of below 6.1% (43 mmol/mol) before conceiving. During pregnancy, she should aim for fasting blood glucose levels of 3.5–5.9 mmol/L and 1-h post-prandial levels of below 7.8 mmol/L.

3. Mrs TM was taking the appropriate dietary supplement; however, the recommended dose for women with diabetes is 5 mg daily, rather than 400 μcg daily. The 5-mg strength tablets are available on prescription.

4. Mrs TM should have been advised to stop her ramipril and simvastatin since both have been associated with an increased risk of birth defects.

Case 44.2

Mr LG is a 47-year-old man with type 2 diabetes. He has recently had basal insulin (insulin detemir) added into his other diabetes medicines: metformin modified release 1 g twice a day and gliclazide 80 mg twice a day. He complains of waking with a headache and feeling ‘groggy’ and unrested in the morning. His recent blood glucose readings have generally been very good although his before breakfast readings are 10–13 mmol/L. He is worried because he is feeling worse since he started insulin, even though his blood glucose levels are much improved. He has made an appointment with his primary care doctor. His primary care doctor suspects nocturnal hypoglycaemia may be causing his recent symptoms.

Questions

1. What is nocturnal hypoglycaemia?

2. Why might nocturnal hypoglycaemia cause raised blood glucose levels in the mornings?

3. How can the diagnosis of nocturnal hypoglycaemia be confirmed?

4. How should it be treated?

Answers

1. Nocturnal hypoglycaemia is a low blood glucose reading that occurs during the night. Definitions vary but it is generally accepted that a reading of less than 3.5 mmol/L is ‘hypoglycaemia’. The normal symptoms of hypoglycaemia may be missed if the patient does not wake. However, signs noticed (often by partners) might include restlessness, sweating and nightmares. Symptoms experienced by the patient in the mornings commonly include headache, lethargy and raised blood glucose levels.

2. Nocturnal hypoglycaemia may be caused by a rebound rise in blood glucose levels due to the ‘somogyi effect’. This is the effect of the counter-regulatory hormones, glucagon, cortisol, adrenaline (epinephrine) and growth hormone, which all increase glucose in response to low levels.

3. Commonly, nocturnal hypoglycaemia can be confirmed by undertaking a blood glucose reading in the early hours of the morning, that is, between 2 and 4 am. This may require the patient to set their alarm to be woken at this time.

4. Nocturnal hypoglycaemia once confirmed may be treated by either having a bed-time snack, or by reducing the dose of night-time insulin. In the case of Mr LG, stopping his sulphonylurea may also help resolve the problem. Care must be taken when interpreting raised blood glucose levels in the morning, since the obvious intervention would be to increase the dose of night-time insulin. However when nocturnal hypoglycaemia is the cause, increasing the insulin dose would only make the problem worse and more dangerous for the patient.

Case 44.3

Mr PT is a 69-year-old man with longstanding type 2 diabetes. He has recently noticed that his right shoe has been rubbing his foot, which he finds confusing since he has been wearing these shoes for 6 months with no problems. His whole left foot now looks red and swollen and when Mr PT inspected it closely, he noticed that there was a weeping sore. However, his foot is not painful, so he does not feel too concerned.

Questions

1. What is the most likely reason that Mr PT did not feel any pain associated with the sore?

2. Why might Mr PT’s shoe suddenly have started to rub his foot?

3. Should he be more concerned?

Answers

1. Mr PT has probably developed sensory neuropathy in his feet. This usually begins with the loss of the sensation of vibration and then may progressively lead onto the loss of sensation altogether.

2. The shape of the feet of people with diabetes has been observed to change over time. This may be due to the development of neuropathy which can weaken muscles causing alterations to the shape of the arch of the foot and toes. In this case, we know that Mr PT has sensory neuropathy since he is unable to feel the pain of the weeping sore. It is likely that Mr PT may also have motor neuropathy.

3. Mr PT should be concerned because his lack of sensation does not indicate that the foot injury is not serious. He is at risk of developing an infected ulcer and needs prompt treatment to prevent the problem from becoming severe. If he does not seek treatment, he may even risk losing his foot through amputation.

Case 44.4

Miss IL is a 17-year-old teenager with recently diagnosed type 1 diabetes. She has been admitted to hospital with diabetic ketoacidosis (DKA), which was precipitated by a diarrhoea and vomiting bug. As she was vomiting she was not eating, and hence she temporarily stopped injecting her insulin. She is normally well controlled on a basal bolus insulin regimen comprising insulin glargine (Lantus) at night and insulin glulisine (Apidra) three times daily with meals.

Questions

1. Why was it a mistake for Miss IL not to inject her insulin whilst she was not feeling well enough to eat?

2. What advice should patients on insulin be given regarding glucose management when they are feeling unwell and not able to eat normally?

3. What are the initial management priorities for patients admitted with diabetic ketoacidosis?

4. What is the correct way for subcutaneous insulin to be re-introduced after a patient has been on a continuous intravenous insulin infusion?

Answers

1. This is a common misunderstanding amongst patients and sometimes even health care professionals. When a person is unwell, their basal insulin requirements can often increase, despite not eating. This is because of the stress involved and the increase in the production of counter-regulatory hormones which increase glucose levels.

2. Patients should be counselled on what are commonly referred to as ‘sick day rules’ and should be given contact numbers for advice if they are unclear or struggling with management. Generally patients need to increase the frequency of glucose monitoring to 4-hourly or more and test urine for ketones on a regular basis. Carbohydrate intake should be maintained as much as possible using sugary drinks or fruit juice, soups, jelly or snack foods if they are having difficulty in eating. Fluid intake is important and patients should be advised to have a glass of water every hour, aiming for 3 L in 24 h. If the blood glucose level is less than 12 mmol/L and ketone result is negative-small, patients should continue their normal insulin dose. If the blood glucose level is between 12 and 18 mmol/L and the ketone result is negative-small, the patient should re-test within 4 h or at the next meal and add 4 extra units of insulin to doses of up to 20 units and 8 extra units to doses greater than 20 units. If the blood glucose level is more than 18 mmol/L and ketones are moderate or large, patients should take 50% extra insulin. If blood glucose is more than 20 mmol/L and ketones are moderate or large, the dose of insulin should be doubled. If the patient is vomiting and blood glucose levels are over 16 mmol/L, then medical help should be sought, especially if there is no improvement after the second test or the next insulin injection.

3. Intravenous sodium chloride 0.9% should be started as soon as possible. A fixed rate intravenous insulin infusion should then be started. Current recommendations are to begin at a rate of 0.1 units/kg. Regular hourly monitoring of blood glucose and ketones should be undertaken and 2-hourly monitoring of serum potassium for the first 6 h.

4. When the patient is biochemically stable, they may be converted back to subcutaneous insulin. The current national guidelines from the Joint British Societies Inpatient Care Group (2010) recommend that long-acting subcutaneous insulin analogues are continued throughout treatment of diabetic ketoacidosis in order to prevent rebound hyperglycaemia when the intravenous infusion is stopped. Therefore, Miss IL’s glargine should have been continued. Her rapid-acting insulin (Apidra®) should be re-introduced at the next meal and the intravenous insulin infusion should be stopped 30 min afterwards.

References

DAFNE Study Group. Training in flexible, intensive insulin management to enable dietary freedom in people with type 1 diabetes: dose adjustment for normal eating (DAFNE) randomised controlled trial. Br. Med. J.. 2002;325:746-749.

Davies M., Storms F., Shutler S., et al. Improvement of glycaemic control in subjects with poorly controlled type 2 diabetes. Diabetes Care. 2005;28:1282-1288.

Department of Health. National Service Framework for Diabetes: Standards. London: Department of Health, 2002.

Dormandy J.A., Charbonnel B., Eckland D.J.A., et al. Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone Clinical Trial In macroVascular Events): a randomised controlled trial. Lancet. 2005;366:1279-1289.

Holman R., Paul S., Bethel M., et al. 10-year follow-up of intensive glucose control in type 2 diabetes. N. Engl. J. Med.. 2008;359:1577-1589.

Holman R., Farmer A., Davies M., et al. Three-year efficacy of complex insulin regimens in type 2 diabetes. N. Engl. J. Med.. 2009;361:1736-1747.

Joint British Societies Inpatient Care Group. The Management of Diabetic Ketoacidosis in Adults. NHS Diabetes; 2010. Available at: http://www.diabetes.nhs.uk/

Khan R., Buse J., Ferrannini E., et al. The metabolic syndrome: time for a critical appraisal. Joint statement from the American Diabetes Association and the European Association for the Study of Diabetes. Diab. Care. 2005;28:2289-2304.

National Institute for Health and Clinical Excellence. Type 1 Diabetes: Diagnosis and Management of Type 1 Diabetes in Children, Young People and Adults, Clinical Guideline 15.. London: NICE. 2004. Available at: http://www.nice.org.uk/nicemedia/pdf/CG015NICEguideline.pdf

National Institute for Health and Clinical Excellence. Obesity: Guidance on the Prevention, Identification, Assessment and Management of Overweight and Obesity in Adults and Children, Clinical Guideline 43.. London: NICE. 2006. Available at: http://www.nice.org.uk/nicemedia/pdf/CG43NICEGuideline.pdf

National Institute for Health and Clinical Excellence. Continuous Subcutaneous Insulin Infusion for the Treatment of Diabetes Mellitus, Technology Appraisal 57.. London: NICE. 2008. Available at: http://www.nice.org.uk/nicemedia/live/12014/41300/41300.pdf