Endocrinology: Diabetes mellitus

Overview

Diabetes mellitus is a metabolic disorder characterised by persistent hyperglycaemia which is a result of defects in
insulin secretion, insulin action or both. The diagnosis and monitoring of diabetes is outlined in Chapter 43 and diabetes is broadly classified as type 1 or type 2 (Chapters 45–48) although other rarer causes of diabetes exist, including genetic disorders (Chapters 44 and 59).
Diabetes is a multisystem disease that carries significant morbidity and mortality from its chronic macrovascular and
microvascular complications (Chapters 49–51). Throughout the course of ‘living with diabetes’ a number of acute complications can occur (Chapters 52–56) which require careful patient education and management. Furthermore, different life stages, from young people, pregnancy (Chapter 58) to elderly care, require the expertise of the whole diabetes multidisciplinary team (MDT) (Chapter 60).
In 2012, 382 million adults worldwide were estimated to have diabetes and it is thought that 550 million will have diabetes before 2030. Currently, there are approximately 3 million people with diagnosed diabetes in the UK and around 850 000 people with undiagnosed diabetes. In the UK, around 10% of the health service budget is spent on diabetes care and diabetes-related problems. Diabetes therefore carries enormous health burdens and economic implications that will continue to grow as its prevalence increases.

History

Diabetes comes from the Greek ‘to pass through’, and mellitus from the Latin word meaning ‘sweetened with honey’. Ancient Egyptians described features similar to diabetes mellitus around 3000 years ago but the actual term ‘diabetes’ was only first used by the physician Aretaeus of Cappadocia in the 2nd century AD. Later, in 1675, ‘mellitus’ was added by Thomas Willis, a physician who re-discovered the urine’s sweet taste. A major turning point in the history of diabetes
was the discovery and use of insulin by Banting and Best in 1921 (Figure 42.2). The first oral hypoglycaemic agents were
marketed in 1955.

Landmark studies

Two early and important diabetes studies are the United Kingdom Prospective Diabetes Study (UKPDS) and the
Diabetes Control and Complications Trial (DCCT). UKPDS ran for 20 years from 1977 and showed that intensive therapy
to lower glucose levels in type 2 diabetes was associated with a reduction in microvascular complications. DCCT ran from 1983 to 1993 and tested the value of intensified versus conventional control in patients with type 1 diabetes mellitus (T1DM). The study demonstrated a dramatic benefit of intensified glucose control in primary prevention of retinopathy, kidney disease and neuropathy.

Advances in care

Since the discovery of insulin in 1921, there have been vast advances in diabetes management. Not only has there been a huge development in insulin preparations and delivery systems, but also a growth in the classes of non-insulin agents available to treat diabetes (Chapters 46 and 48) In addition, the pattern of care provision has evolved in the UK. With a growth in the prevalence of diabetes, and the emerging burden of its complications, management has taken on more of a multidisciplinary approach. Diabetic specialist nurses (DSNs) became more common from
the 1980s, when new strengths of insulins and self-monitoring emerged. Diabetes centres based at hospitals helped to establish this MDT approach and acted as a link between primary and secondary diabetes care. The recent move has been towards diabetes management in the community, with a suggestion that secondary care should manage the ‘super six’ in diabetes: pregnancy, diabetic foot care, nephropathy, insulin pumps, inpatient care and T1DM (poor glycaemic control in young people).
National policy has also helped to shape diabetes care in the UK in recent years, including the National Service Framework of standards of care for people with diabetes, the emergence of NICE guidance on diabetes management (www.nice.org.uk) and, more recently, Joint British Diabetes Society guidelines for inpatient diabetes.

Diabetes support organisations

Support for patients with diabetes and their families, as well as providing professional support, has grown over the past decades in the UK. In 1934, the Diabetic Association was set up which later became the British Diabetic Association and then Diabetes UK (www.diabetes.org.uk) at the turn of the millennium. The charity aims to promote the study, knowledge and treatment of diabetes in the UK. Its local voluntary groups provide support and information to people with diabetes across the UK.
For clinicians in the UK managing diabetes, the Association of British Clinical Diabetologists (ABCD) (www.diabetologistsabcd. org.uk) exists to promote care for patients with diabetes among specialists, and acts as a national platform for training, research and information in diabetes management in the UK.

Awareness

Diabetes publicity campaigns, better education, increased media coverage and the explosion of social networking have increased public and professional awareness of diabetes but much more needs to be done.
Since 2006, the universal symbol for diabetes has been the blue circle. The symbol aims to raise awareness about diabetes, inspire new activities, bring diabetes to the attention of the general public, brand diabetes and provide a means to show support for the fight against diabetes.

Diagnosis and monitoring

Diagnosis of diabetes

The diagnosis of diabetes is made either in the light of symptoms or on routine ‘screening’ (Box 43.1). The clinical presentations of T1DM and T2DM are discussed in Chapters 45 and 47.

Glucose

unexplained weight loss), the diagnosis can be made based upon the WHO 2006 criteria:

• A random venous plasma glucose concentration ≥11.1 mmol/L,
  or
• A fasting plasma glucose concentration ≥7.0 mmol/L (whole
  blood ≥6.1 mmol/L), or
• Two-hour plasma glucose concentration ≥ 11.1 mmol/L after
  75 g anhydrous glucose in an oral glucose tolerance test (OGTT).
  Diabetes should not be diagnosed in individuals with no symptoms based on a single glucose reading; this requires
  confirmatory testing. At least one additional glucose test result on another day with a value in the diabetic range is essential, either fasting, from a random sample or from the 2-hour post glucose load. If the fasting or random values are not diagnostic, the 2-hour value should be used.

HbA1c

More recently, glycosylated haemoglobin (HbA1c) has been introduced as a method for diagnosing diabetes. HbA1c is
formed by glycation of haemoglobin as it is exposed to plasma glucose and reflects the beta-N-1-deoxy fructosyl element of haemoglobin. HbA1c reflects average plasma glucose over the previous 8–12 weeks. It can be performed at any time of the day and does not require any special preparation such as fasting. HbA1c can be expressed as a percentage (DCCT unit) or as a value in mmol/mol (IFCC unit). The latter has been adopted in the UK since 2009.
In 2011, the WHO recommended an HbA1c of 48 mmol/mol (6.5%) as the cut-off point for diagnosing diabetes. When HbA1c is ≥48 mmol/mol (6.5%), diagnosis should be confirmed with a second sample, unless the individual is symptomatic with plasma glucose levels ≥11.1mmol/L, when confirmation is not needed. If the second sample is <48 mmol/mol (6.5%), the patient should be treated as at high risk of diabetes and the test should be repeated in 6 months or sooner if symptoms develop. A value of <48 mmol/mol (6.5%) does not exclude diabetes. These patients may still fulfill WHO glucose criteria for the diagnosis of diabetes, hence glucose testing as described above can be used in patients who have symptoms of diabetes or clinically are at very high risk. However, the use of such glucose tests is
not recommended routinely in this situation. Patients with ‘high normal’ HbA1c levels below the threshold for diabetes diagnosis, particularly ≥42 mmol/mol (6.0%), should receive lifestyle interventions in an attempt to delay and prevent the onset of diabetes. Although HbAlc is an accurate and precise measure of chronic glycaemic levels, there are certain situations when it should not be used in diagnosis:

• Children and young people
• T1DM
• Symptom onset within 2 months
• Pregnancy
• Medications (e.g. steroids) that can cause a rapid rise in glucose
• Genetic, haematologic and illness-related factors that influence HbA1c, such as those with haemolytic anaemia and haemoglobinopathies.

Oral glucose tolerance test

The OGTT is performed by asking the patient to fast for at least 8 hours, usually overnight, and attend for a fasting plasma glucose sample (Figure 43.1a). They are then given a drink containing 75 g anhydrous glucose (e.g. Polycal©) and a further blood sample is taken after 2 hours.

Impaired fasting glycaemia and impaired glucose tolerance

Impaired glucose tolerance (IGT; Figure 43.1b) refers to a fasting plasma glucose <7.0 mmol/L and a 2-hour plasma glucose ≥7.8 but <11.1 mmol/L on an OGTT. Impaired fasting glycaemia (IFG) relates to a fasting plasma glucose of 6.1–6.9 mmol/L (and a 2-hour glucose of <7.8 mmol/L, if measured). Patients in these groups are at higher risk of developing overt diabetes and should be educated regarding lifestyle measures in an attempt to delay or halt its onset.
They should be under regular surveillance to monitor their glucose status, with repeat blood testing at least every 1–2 years.

Screening

Those at higher risk for developing type 2 diabetes should be offered screening. If between the age of 40 and 75, a risk
assessment should be made, and if symptoms of diabetes or risk factors are present (overweight or obese, atherosclerotic disease, a first degree relative with T2DM, or African-Caribbean, Middle Eastern or South Asian origin), testing with a fasting plasma glucose or HbA1c should be offered. It is worth remembering that the threshold for screening should be lower in higher risk groups. Screening should therefore be considered in those above age 25 of South Asian, Chinese, African-Caribbean and black African origin who have a BMI >23 kg/m². Screening for diabetes
during pregnancy is discussed in Chapter 58. Other ‘at-risk’ groups who should be screened include those with known IFG and IGT, women who have had gestational diabetes but have tested normal following delivery and obese women with PCOS.

Monitoring diabetes

The main way of monitoring glycaemic control in diabetes is through measurement of HbAlc (Figure 43.1c). This should be performed at the annual review and at more regular intervals (but not usually less than 2–3 monthly) if glycaemic control needs attention. In those with haemoglobinopathies, fructosamine may be a suitable alternative.
Patients are also encouraged to self-monitor their diabetes using capillary blood glucose monitors (Figure 43.2). Patients should keep a record of these readings in a diary, which can then be reviewed by their diabetes team to note any patterns in their blood glucose readings over days and weeks. This is particularly helpful for those who inject insulin, in titrating their doses of insulin according to blood glucose levels. Some blood glucose monitors are now
available that allow a sensor to be placed under the skin, which is changed periodically, to monitor glucose levels and communicate the results with a hand-held device to show the reading.

Continuous glucose monitoring

In some patients who have problematic control and more information is needed about their glucose patterns, particularly nocturnal fluctuations, a continuous glucose monitoring system can be worn. This is a device that is placed subcutaneously and worn from 24 hours to several days to help note patterns in blood glucose variation, which can be helpful in altering insulin doses or the settings of those on insulin pump therapy.

Classification

Diabetes mellitus has a number of causes and can therefore be classified according to aetiology (Figure 44.1).

Type 1 diabetes

This accounts for 5–10% of diabetes and is autoimmune in aetiology (Chapter 45). This cellular-mediated process results in destruction of the β-cells of the pancreas and absolute insulin deficiency, with patients requiring insulin to survive. It has multiple predisposing genetic and environmental factors that are still not completely understood. Various viruses have been implicated in β-cell destruction but their exact contribution to pathogenesis is unclear. T1DM can be associated with other autoimmune diseases, such as Addison’s disease (Chapter 20), Graves’ disease (Chapter 10), Hashimoto’s disease (Chapter 13) and pernicious anaemia.
A small proportion of patients with T1DM, mainly African or Asian in ethnicity, do not appear to have underlying
autoimmunity. There is usually a strong family history, but no evidence of β-cell destruction and no human leucocyte antigen (HLA) association. These patients have episodic ketoacidosis and varying degrees of insulin requirements.

Type 2 diabetes

Around 90–95% of patients with diabetes have T2DM, caused by both insulin resistance and a defect in insulin secretion (Chapter 47). It is often associated with obesity, particularly abdominal adiposity, and the risk increases with increasing BMI, age and a lack of physical activity. Those with dyslipidaemia, hypertension or with a history of gestational diabetes are also at increased risk, as well as those in certain ethnic groups, including South Asian and African-Caribbean. The genetic predisposition is stronger in T2DM than T1DM, but is polygenic in origin and less clearly understood.

Maturity-onset diabetes of the young

Maturity-onset diabetes of the young (MODY) is associated with monogenic defects in β-cell function with few or no defects in insulin action (Chapter 59). It is inherited in an autosomal dominant fashion and is often characterised by hyperglycaemia at a younger age, usually below the age of 25 years. The most common form is caused by a mutation in a hepatic transcription factor encoded on chromosome 12, referred to as hepatocyte nuclear factor 1α (HNF-1α). Mutations in other genes (e.g. glucokinase) result in other forms of MODY.

Other genetic defects in β-cell function

Point mutations in mitochondrial DNA result in diabetes and deafness, with the most common arising at position 3243 in the tRNA leucine gene. Another defect, inherited as an autosomal dominant condition, impairs the conversion of proinsulin to insulin, resulting in mild glucose intolerance.

Genetic defects in insulin action

Previously known as type A insulin resistance, mutations in the insulin receptor can result in hyperinsulinaemia and a spectrum of hyperglycaemia from mild through to overt diabetes mellitus. Childhood syndromes exist with mutations in the insulin receptor gene, characterised by marked insulin resistance and hyperinsulinaemia. Rabson–Mendenhall syndrome is associated with teeth and nail abnormalities whereas leprechaunism is usually fatal in infancy. Lipoatrophic diabetes is thought to be caused by a defect in the post-insulin receptor signal transduction pathway.

Pancreatic diseases

Any disease process that causes extensive damage to the pancreas can result in diabetes. Pancreatitis (particularly chronic, with multiple insults), infection, trauma, pancreatectomy, haemochromatosis and cystic fibrosis are potential causes.

Endocrine disorders

Various endocrine conditions such as acromegaly, Cushing’s syndrome and glucagonoma can cause diabetes, because of the presence of excessive GH, cortisol and glucagon, respectively, which have insulin-antagonising effects. Patients with these conditions should be tested for diabetes at diagnosis and during the course of their disease.

Drugs

A number of drugs are associated with glucose dysregulation and the development of diabetes. The most common are exogenous steroids, which promote gluconeogenesis and cause insulin resistance. Other drugs affecting insulin secretion can unmask diabetes in those who are already insulin-resistant.

Gestational diabetes

Diabetes develops during some 7% of pregnancies, a condition known as gestational diabetes (Chapter 58).

Other associations with diabetes

Some syndromes predispose to the development of diabetes, including Down’s syndrome, Turner’s syndrome, Kleinfelter’s syndrome and Laurence–Moon–Biedl syndrome. DIDMOAD syndrome, otherwise known as Wolfram’s syndrome, is an autosomal recessive condition characterised by insulin deficiency. Rare conditions can also cause immune-mediated diabetes. Stiff person syndrome is one such autoimmune condition of the CNS characterised by muscle stiffness and spasms, in whom around one-third of patients will develop diabetes associated with
glutamic acid decarboxylase (GAD) autoantibodies. Antibodies to the insulin receptor, resulting in blocking the action of insulin at its receptor site, can also rarely cause hyperglycaemia and diabetes. These antibodies are sometimes found in patients with other autoimmune conditions, such as systemic lupus erythematosus.

Type 1 diabetes: aetiology and clinical presentation

Aetiology

T1DM is an autoimmune disease with both genetic and environmental factors playing an important part in its
development. Chapter 59 explores the genetics of diabetes in more detail. Genetic factors are thought to account for around 30% of the susceptibility risk.

Genes

The risk of developing T1DM is 0.4% in the general population, 1–2% if the individual’s mother has T1DM, 3–5% if the father has diabetes, with up to 35% concordance in monozygotic twins. Genes in the major histocompatibility complex (MHC) antigens/ HLA glycoprotein molecule system are involved in disease susceptibility. HLA class II molecules bind foreign antigen peptides and present them to T-helper lymphocytes. HLADR- 3-DQ2/DR-4-DQ8 class II HLA antigens are found in over 95% of Europeans with T1DM.

Environment

Environmental factors are thought to act as triggers for autoimmunity (Figure 45.1a). A number have been proposed,
including viruses (mumps, rubella, cytomegalovirus), bacteria, stress, intrauterine factors (maternal rubella, pre-eclampsia, birth weight) and dietary factors. It is postulated that such factors lead to upregulation of HLA-antigens in genetically predisposed individuals, or exposure to an infective trigger can lead to the presentation of self-antigens to T-helper cells.

Pathophysiology

The autoimmune process, involving both humoral and cellular immunity, results in CD8 T-cell lymphocyte-mediated
destruction of the insulin-secreting β-cells (Figure 45.1). The chronic inflammatory changes which ensue include infiltration with CD4+ and CD8+ lymphocytes and macrophages, causing an insulinitis. β-Cell destruction subsequently occurs, with a loss in β-cell mass and consequent insulinopenia. In the absence of insulin action in muscle and adipose tissue, glucose is not transported into the cells by the GLUT4 transporter. The clinical manifestations of T1DM appear as a result.
A number of islet-related antibodies are present in patients with T1DM, which can be present for many months before the clinical onset of disease (Figure 45.1b). The islet autoantibodies GAD and islet antigen 2 (IA-2) are present in up to 90% of patients with newly diagnosed T1DM.

Clinical presentation

Symptoms of T1DM usually develop over a short period, typically over 1–4 weeks. Patients are generally younger than those with T2DM, with a peak onset at age 12 years.

Osmotic symptoms

The most common symptoms are those of thirst, polydipsia, polyuria and weight loss (Figure 45.2). Hyperglycaemia results in a marked osmotic effect, often more severe than T2DM. The increased osmotic effect can lead to profound dehydration, hypovolaemia and drowsiness. Hyperglycaemia causes osmotic changes in the lens of the eye, with subsequent blurred vision. In addition, a hyperglycaemic environment predisposes patients to cutaneous Candida infections, particularly genital thrush.

Catabolic symptoms

Absolute insulin deficiency also results in protein breakdown and muscle wasting, fatigue and weight loss.

Acute presentation

A patient with new onset disease can also present in diabetic ketoacidosis (DKA), a diabetic emergency (Chapter 52).
Approximately 25% of children with T1DM present in DKA.

Type 1 diabetes: insulin and other therapies

Lifestyle

All patients with T1DM should be offered lifestyle advice. Dietary advice should cover the hyperglycaemic effects of different foods in the context of insulin therapy, the effects of different foods on glycaemia, the place of snacks between meals and at bedtime, healthy eating to reduce arterial risk and the effect of alcohol-containing
drinks. Advice should also be given on physical activity, including the appropriate intensity and frequency, the role of self-monitoring and insulin dose adjustment around exercise and the effects on glycaemia. Smoking cessation strategies should be discussed with smokers.

Insulin

The main principle of insulin therapy is to replace insulin in a way that follows the normal physiological pattern of secretion as closely as possible (Figure 46.1a).
Three main types of insulin are available:
1 Soluble insulin These are administered subcutaneously, or intravenously (e.g. during acute diabetic emergencies).
2 Protamine insulin/zinc suspensions (isophane insulins) These act as a basal insulin, with a prolonged insulin action.
3 Insulin analogues Rapid-acting analogues (insulin lispro, aspart) are more rapidly absorbed than soluble insulin. Long-acting analogues (insulin glargine or insulin detemir) provide a stable basal concentration of insulin.
There are two common insulin regimens used in patients withT1DM:
1 Twice daily insulin regimens, comprising a twice daily injection of pre-mixed insulin (a combination of short- and
intermediate-acting insulin) given before breakfast and evening meal. Twice daily frequency is an advantage but it leaves little flexibility in the timing and size of meals, and carries a higher risk of hypoglycaemia. This regimen generally suits those with a fixed eating pattern or who need assistance with injecting insulin, such as those with learning difficulties.
2 Multiple daily injections (MDI)/basal bolus regimen comprises a once daily basal insulin (isophane [NPH] or analogue
[glargine or detemir]) in combination with short-acting soluble or analogue insulin given at mealtimes or with snacks. This allows more flexibility in the timing and quantity of meals, reduces the risk of hypoglycaemia and facilitates better glycaemic control.

Education

resuspension of insulin, use of insulin pens (Figure 46.1b), injection techniques and insulin dose adjustment. Injection sites should be checked regularly to ensure there is no development of lipohypertrophy, a build-up of subcutaneous fat at the sites, which can result in variable insulin absorption. Rotating the site of injection within a particular area should be encouraged to avoid this (Figure 46.1c).
Structured education programmes should be offered to all patients with T1DM. The ‘Dose Adjustment For Normal Eating’ (DAFNE) is a national programme that provides patients with education including carbohydrate counting (adjusting meal boluses of insulin to match the carbohydrate intake). Patients attending such courses have been shown to have improved quality of life and glycaemic control.

Home glucose monitoring

Blood glucose monitoring is essential for day-to-day management. Hand-held capillary blood glucose monitors allow patients to measure their glucose level with a finger-prick blood test. This is usually monitored fasting (on waking), and immediately before a meal or around 1–2 hours after a meal, several times a day. Patients are encouraged to keep a diary of their blood glucose recordings, although meters now have built-in functions to enable readings to be downloaded and viewed electronically. Meters are also available that record glucose readings from a small subcutaneous sensor which allow the reading to be taken by scanning the meter over the sensor. Relevant information should be provided to patients receiving insulin regarding driving, travel, leisure activities and work. Long-term control of glycaemic control is monitored by HbA1c.

Continuous subcutaneous insulin infusion

Continuous subcutaneous insulin infusion (CSII), otherwise known as ‘insulin pump’ therapy, is currently available for patients in the UK who fail to achieve adequate glycaemic control on an MDI regimen without experiencing disabling hypoglycaemia or those who have an HbA1c ≥69 mmol/mol despite being on an MDI regimen with a high level of educational input. The insulin pump devices consist of an insulin reservoir containing short-acting insulin that is continuously infused into subcutaneous tissue (Figure 46.1d). The basal rates can be set and altered for different periods of the day, while boluses can then be given at mealtimes. CSII can improve glycaemic control and reduce hypoglycaemia in well-motivated individuals.

Pancreatic transplantation

Allogenic pancreatic islet cell transplantation is a procedure in which islet cells are retrieved from pancreases of brain-dead donors (Figure 46.1e). Under local anaesthesia, cells are inserted percutaneously into the portal vein and infused into the liver. Sometimes more than one infusion is required. It is generally indicated for those who have recurrent severe hypoglycaemic episodes or who have lost hypoglycaemic awareness, or with suboptimal diabetes control already on immunosuppressive therapy after a renal transplant. Although a reduction in severe hypoglycaemic episodes is seen following the procedure and insulin independence can occur in up to 60% patients at 1 year,
ongoing immunosuppression is required and only 10–20% remain insulin-free at 5 years. Low dose insulin therapy is usually required for most in the long term. Whole organ pancreas transplant is an alternative method which can result in 55% of patients being insulin-free at 5 years. However, it involves risks associated with surgery and requires
immunosuppressive treatment for as long as the transplant continues to work.

The artificial pancreas

The artificial pancreas is a system, worn like an insulin pump, that measures blood glucose levels on a minute-to-minute basis using a continuous glucose monitor (CGM), and transmits this information to an insulin pump that calculates and releases the required amount of insulin into the body. Although still in the trial stage, the technology appears to improve the time spent in normoglycaemia, reducing the frequency of hypo- and hyperglycaemic episodes.

Immunotherapy

Immunotherapy is emerging as a potential future therapy in targeting the autoimmune islet cell destruction in T1DM. The aim is to slow down or prevent the disease process, and several trials are ongoing.

Type 2 diabetes: aetiology and clinical presentation

Aetiology

Type 2 diabetes (T2DM) classically presents over the age of 40, although in high risk populations, such as South Asian, African and African-Caribbean ethnicities, it can present much earlier. Age and ethnicity are risk factors for T2DM, with the percentage doubling over the age of 65 years and with a sixfold increase in prevalence in high-risk ethnic groups (Figure 47.1a). In recent years, T2DM has increasingly been diagnosed in childhood, largely because of physical inactivity and obesity. The first cases of T2DM diagnosed in children were identified in those of Pakistani, Indian or Arabic origin. Children of South Asian origin are 13 times more likely to develop T2DM than Caucasians.
Genetic factors are thought to have a significant role in the aetiology of T2DM, accounting for up to 80% of disease
susceptibility. It is a polygenic disease, with no single gene defect being responsible for its development. Chapter 59 further explores the genetics of diabetes.
Environmental factors are also important in the development of insulin resistance and T2DM. Obesity is a key factor,
particularly visceral (central) adiposity, and can be established clinically by an increased waist circumference. Physical exercise is also an important factor, with an increased risk in those with a sedentary lifestyle. Another emerging risk is the intrauterine environment, with both low and high birth weight associated with insulin resistance.

Pathophysiology

T2DM is characterised by a defect in both insulin sensitivity and insulin secretion.
Insulin resistance occurs at the level of the peripheral tissues (skeletal muscle, adipose tissue) and liver, resulting in reduced glucose uptake in skeletal muscle and impaired inhibition of hepatic glucose output. In adipose tissue, insulin resistance leads to increased non-esterified fatty acid production, which stimulates gluconeogenesis and triglyceride synthesis. It is thought that an insulin signalling defect underlies insulin resistance in T2DM, particularly down-regulation of post-receptor signalling. As a result of β-cell dysfunction, insulin secretion is already reduced by half by the time of diagnosis of T2DM, a process that can begin up to 10 years before presentation (Figure 47.1b). This
results in a reduction in the initial first-phase insulin response to glucose challenge. Genetic and environmental factors are thought to contribute to this, including obesity, glucose and lipid toxicity.
Patients passing from IGT to T2DM are characterised by rising insulin resistance, initial compensatory hyperinsulinaemia to maintain glucose concentrations within the normal range, but eventual β-cell exhaustion, resulting in a rise in glucose levels.

Clinical presentation

One-third of cases are detected incidentally, often on a routine screening blood test or following a cardiac event.
As the development of the disease is slow, diagnosis is often delayed for many years. The patient can therefore present with complications from prolonged hyperglycaemia, including microvascular complications such as peripheral neuropathy or diabetic retinopathy, or with recurrent infections (Figure 47.1c). Some 10% of individuals presenting with T2DM have established microvascular complications at the time of diagnosis. Only about half of patients present with the classic symptoms of thirst, polydipsia, polyuria and tiredness secondary to hyperglycaemia, although these symptoms are often less marked than in T1DM. Weight loss is an unusual feature at presentation. Up to 25% of patients present as an emergency in a hyperglycaemic hyperosmolar state (Chapter 53).

Type 2 diabetes: treatment

Treatment of T2DM can broadly be divided into lifestyle (diet and exercise) and pharmacological treatments (Figure 48.1). The HbA1c target should be individualised, and may well need to be above 48 mmol/mol (6.5%) depending on the patient’s circumstances. For example, a more relaxed target may be appropriate in elderly patients with recurrent hypoglycaemia.

Diet and exercise

Dietary modifications and exercise improve insulin sensitivity and glycaemic control. A balanced ‘healthy’ diet should be
advised, with reduced amounts of refined sugars and saturated fats, and increased proportions of complex carbohydrates and fibre. Thirty minutes of exercise a day should also be encouraged, particularly exercises that the patient is willing and able to maintain long term. Lifestyle modification is the recommended initial approach for most patients with T2DM but HbA1c should be monitored at 3 months, and pharmacological therapy considered if above target. A step-wise approach is subsequently adopted.

Metformin

Metformin has been in use for over 50 years, and remains the first line drug for most patients. It belongs to the biguanide group of drugs and acts as an insulin sensitiser by increasing glucose uptake in skeletal muscle and adipocytes, reducing 115 hepatic gluconeogenesis and glycogenolysis, and reducing glucose absorption from the small bowel. It has a small effect on weight loss, reduces appetite and has a ‘cardioprotective’ benefit. Gastrointestinal side effects (diarrhoea, abdominal pain and nausea) can occur in 10–20% of patients but can be reduced with modified-release metformin. Metformin is also rarely associated with lactic acidosis, so care must be taken in patients with renal and liver failure. There is an association of metformin with vitamin B12 deficiency but levels should not be routinely
monitored unless deficiency is clinically suspected.

Sulphonylureas

Sulphonylureas (SUs) are insulin secretagogues, and stimulate insulin release from the β-cells by acting on the sulphonylurea receptor. The most commonly prescribed SU in the UK is gliclazide, although glibenclamide, glipizide and glimepiride are also used. Although SUs can result in rapid symptomatic improvement, their main side effects are weight gain and hypoglycaemia. They are typically used second line but can be used first line when
metformin is not tolerated, the patient is not overweight or rapid treatment of symptomatic hyperglycaemia is needed. SUs may not be appropriate in patients where the risk of hypoglycaemia is an important consideration (e.g. certain occupations).

Meglitinides

These drugs, nateglinide and repaglinide, are less commonly used in clinical practice. They stimulate insulin release in the early post-prandial phase but have modest benefits on overall glycaemic control. Hypoglycaemia can occur but, given their shorter duration of action, this is less marked than with SUs.

Acarbose

Acarbose is an α-glucosidase inhibitor, an enzyme found in the brush border of the small intestine, which digests carbohydrates. It therefore acts to slow dietary carbohydrate breakdown, reduce intestinal glucose uptake and the subsequent post-prandial glucose peak. The effects on HbA1c are less impressive than with other oral agents hence its use in practice is limited.

Glitazones

Thiazolidinediones (TZDs or ‘glitazones’) are insulin sensitisers that work by binding to peroxisome proliferator-activated receptor gamma, resulting in increased expression of glucose transporter 4, causing improved glucose and fatty acid uptake, particularly in adipose tissue. Insulin sensitivity is thus improved by reduced availability of fatty acids to muscle. The main side effect is weight gain, related in part to fluid retention, hence care should be taken when used in patients with heart failure. Rosiglitazone has been removed from the market because of its link with myocardial infarction. Pioglitazone is still in use but has been linked with a possible increased risk of bladder cancer and osteoporotic fractures.

DPP-4 inhibitors

These drugs, also known as the ‘gliptins’, are oral therapies that include sitagliptin, saxagliptin, vildagliptin and linagliptin. They belong to the incretin-based group of therapies. The incretin hormones, which include glucagon-like peptide 1 (GLP-1) and glucose-dependent insulinotrophic polypeptide (GIP), are gut hormones secreted in response to eating. These hormones cause glucose-induced insulin secretion, reduce glucagon secretion, delay gastric emptying and reduce satiety. Dipeptidyl peptidase 4 (DPP-4) is an enzyme that breaks down GLP-1 in the gut. The glitpins act to inhibit DPP-4, thus preventing the rapid breakdown of GLP-1, doubling its concentration and that of GIP. They are generally well tolerated, improve HbA1c and are weight-neutral.

GLP-1 receptor agonists

GLP-1 receptor agonists are injectable therapies that are analogues of GLP-1. They improve glycaemic control and can
also lead to significant reduction in weight. Exenatide (injected either twice daily or a slower release preparation given once weekly), liraglutide (once daily) and lixisenatide (once daily) are the main GLP-1 receptor agonists currently in use. Their main side effects are nausea and vomiting, which occur in up to 50% of patients, although this tends to settle with time. Because of a possible association with pancreatitis, GLP-1 receptor agonists should be avoided in patients with a history of, or at risk of pancreatitis.

SGLT-2 inhibitors

Sodium glucose co-transporter 2 (SGLT-2) inhibitors are relatively new drugs that act to inhibit SGLT-2, a co-transporter
found in the proximal tubule of the kidney. This is responsible for re-absorption of up to 90% of glucose filtered through the glomeruli. Urinary glucose excretion is increased, resulting in improved HbA1c and weight loss. Glycosuria can lead to an increased risk of genitourinary tract infections, particularly vaginal thrush and candida balanitis. The main SGLT-2
inhibitors available in the UK are canagliflozin, dapagliflozin and empagliflozin.

Insulin

Insulin is indicated in patients with T2DM when inadequate glycaemic control is achieved on oral agents. It is needed in
around 30–40% patients at an average of 11 years from diagnosis. NPH insulin is recommended first line, either administered once or twice daily. If the patient is at risk of recurrent symptomatic hypoglycaemia then a long-acting insulin analogue (glargine or detemir) is recommended as an alternative. If this fails to achieve adequate glycaemic control, short-acting prandial insulin can be added. Alternatively, a pre-mixed insulin can be used, especially for control of post-prandial hyperglycaemia and in those who are eating at regular times each day.

Macrovascular complications

Epidemiology

Cardiovascular disease (CVD) is the leading cause of mortality in T1DM and T2DM, accounting for up to 80% of global deaths. The risk of myocardial infarction (MI) in patients with diabetes is the same as for a patient without diabetes with a history of previous MI.

Aetiology

Atherosclerosis, the process underpinning CVD, arises as a result of injury and chronic inflammation in the arterial wall, resulting in accumulation of oxidised lipids and low density lipoprotein in the endothelium. The inflammatory response leads to macrophage infiltration, foam cell formation and smooth muscle cell proliferation. The atherosclerotic, lipid-rich lesion that forms can eventually rupture, causing an acute ischaemic event. This process is accelerated in diabetes, whereby hyperglycaemia stimulates production of advanced glycation end-products, leading to inflammation and vasoconstriction. Additional risk factors, including endothelial dysfunction, hypercoagulability, hypertension, dyslipidaemia and central obesity are more prevalent in T2DM and often cluster together in the form of the metabolic syndrome. This accelerates CVD risk.

Clinical presentation

Patients present with symptoms relating to the site of atherosclerosis but these usually occur when disease is already
well-established. Ischaemic pain can be diminished or absent in diabetes so symptoms do not always correlate well with disease extent.
Angina or an acute coronary syndrome can occur in patients with ischaemic heart disease. Cerebrovascular ischaemia can manifest with neurological symptoms (e.g. speech, gait, power, sensory disturbances) in a transient ischaemic attack or stroke. Peripheral vascular disease can present with intermittent claudication, an acute ischaemic limb (pain, pallor, cold, pulseless) or with a new foot ulcer resulting from underlying silent ischaemia (Figure 49.1).
There can be other early symptoms or signs that point to an increased CVD risk. Erectile dysfunction is one such example, as is the presence of microalbuminuria (Chapter 50). Given that symptoms and clinical events often only occur
when atherosclerosis is already well advanced, CVD screening and prevention is a critical part of management.

Management

Lifestyle measures

Exercise and physical activity, in the form of moderate intensity aerobic exercise 5 times a week or 75 minutes a week of vigorous intensity aerobic exercise, should be recommended (Figure 49.2e). Patients should follow a diet in which total fat is 30% or less of total daily intake and saturated fats are 7% or less of total energy intake. Saturated fats should be replaced by monounsaturated and polyunsaturated fats. Alcohol intake should not exceed 3–4 units/day in men and 2–3 units/day in women. Smoking cessation should also be promoted and counselling or nicotine replacement therapy offered.

Blood pressure control

Hypertension occurs in more than 75% of patients. Blood pressure should be measured annually in those with no history of hypertension or renal disease. Target blood pressure should be <140/80 mmHg or <130/80 mmHg in the presence of retinopathy, nephropathy or known CVD. Add-on treatments include calcium channel blockers, thiazide diuretics, alpha-blockers or potassium-sparing diuretics. ACE inhibitors are generally first line therapy, with angiotensin II receptor blockers used in those intolerant of ACE inhibitors. Add-on treatments include calcium channel blockers, thiazide diuretics, alpha-blockers or potassium-sparing diuretics. In women wishing to become pregnant, a calcium channel blocker is recommended first line.

Lipid lowering therapy

In T1DM, statins should be offered for the primary prevention of CVD in those:

• Older than 40 years, or
• With diabetes for more than 10 years, or
• Who have established nephropathy
• With other CVD risk factors.
  A risk assessment for CVD in T2DM can be made using the QRISK2 assessment tool (www.qrisk.org). Statin therapy is recommended for the primary prevention of CVD in patients with T2DM who have a ≥10% 10-year risk.

Glycaemic control

It is well established that hyperglycaemia is a risk factor for CVD, with the UKPDS showing that the incidence of MI rose
by 14% for each 1% rise in HbA1c. However, the role of intensive glucose control in reducing this risk is still unclear, with some studies suggesting that this can actually increase risk. Some drug therapies used to treat diabetes can also have beneficial effects in reducing macrovascular disease. This is especially true for metformin, which been shown to be associated with a decrease in CVD events.

Aspirin

Aspirin is no longer recommended for primary prevention of CVD in all patients with diabetes. However, it may still have a role in those with diabetes at high cardiovascular risk and continues to be used in the treatment of acute coronary syndromes and in secondary prevention.

Interventional therapy

Patients with coronary artery disease may undergo percutaneous coronary angiography and stent insertion for
occlusive coronary lesions. However, atherosclerosis is usually more diffuse in diabetes, hence coronary artery bypass graft surgery may be needed. In patients with peripheral vascular disease, lower limb angioplasty or surgical bypass can be used as interventions.

Microvascular complications

Background

Around 20% of patients with newly diagnosed T2DM have microvascular complications, which reflects the long duration of disease before diagnosis.

Aetiology

Various processes, all driven by hyperglycaemia, are thought to have a role in disease development: advanced glycation end-products (AGEs), reactive oxygen species and cytokines such as vascular endothelial growth factor (VEGF), resulting in cellular damage. Although chronic hyperglycaemia has a major role in pathogenesis, hypertension and activation of the renin– angiotensin system are also important.

Neuropathy

Peripheral neuropathy

Neuropathy can affect up to half of patients with diabetes. The most common form is a distal symmetrical sensorimotor
polyneuropathy, also called diabetic peripheral neuropathy (DPN). A lack of sensation in a ‘glove and stocking’
distribution can be present or patients complain of neuropathic (‘electric-shock’ or ‘burning’) pain, typically worse at night. Mononeuropathies (e.g. affecting the median, ulnar or radial nerves, or cranial nerves III or IV) can also occur but tend to present acutely (Figure 50.1a). DPN puts patients at increased risk of foot disease hence patient education and surveillance is vital. In cases of painful DPN, duloxetine or amitriptyline can be introduced, with tramadol or stronger opiates introduced if pain is not controlled.

Autonomic neuropathy

Autonomic neuropathy can develop in patients with longstanding, poorly controlled diabetes. Any of the autonomic
nerves can be affected, leading to a range of manifestations (Figure 50.1b):

• Cardiovascular: resting tachycardia, postural hypotension, silent ischaemia, sudden cardiac death
• Gastrointestinal: gastroparesis, diarrhoea, constipation, oesophageal dysmotility
• Genitourinary: erectile/bladder dysfunction.
  Autonomic neuropathy is managed by treating symptoms and improving glycaemic control. Gastroparesis can be
  managed with drugs including antiemetics and erythromycin, or, in more severe cases, by gastroelectrical stimulation (a ‘gastric pacemaker’). Erectile dysfunction can be managed with phosphodiesterase-5 inhibitors, such as sildenafil or tadalafil, or if needed with vacuum pumps, alprostadil injections or pellets.

Nephropathy

Around 25–50% of patients will develop nephropathy, which is the most common single cause of end-stage renal disease (ESRD) requiring dialysis or a kidney transplant in the UK. The disease is characterised by an increase in urinary albumin excretion (20–200 μg/min), detected as microalbuminuria (Figure 50.1c). Around 20–30% progress to frank proteinuria, which can even lead to nephrotic syndrome. Glomerular filtration rate (GFR) becomes abnormal when persistent proteinuria has developed. Patients with early nephropathy are asymptomatic; only those with established disease develop clinical features of hypertension, oedema and ultimately uraemic symptoms (nausea, lethargy, poor appetite, itching).
Screening for early disease is therefore critical to slow progression. Annual assessment of urinary albumin : creatinine
ratio should be undertaken, with two out of three abnormal samples required to confirm microalbuminuria. Estimated GFR(eGFR) should also be assessed annually.
Optimal BP control is vital in preventing progression, aiming for a target of <130/80 mmHg. Other cardiovascular risk
factors, such as smoking cessation and lipid lowering, should be managed appropriately. ACE inhibitors or angiotensin receptor blockers are the treatments of choice to control BP and prevent progression of microalbuminura to frank proteinuria and ESRD. Referral to a nephrologist should occur if eGFR falls below 30 mL/min/1.73 m², if there is a rapid decline in eGFR, in the presence of uncontrolled hypertension or unexplained anaemia, or in cases of proteinuria thought to be non-diabetic in origin (e.g. if retinopathy is absent). Renal replacement therapy (RRT) can be considered in ESRD, usually in the form of haemodialysis, continuous peritoneal dialysis or renal transplantation. As with all microvascular complications, good glycaemic control is important: the UKPDS and the DCCT showed that lowering
HbAlc by 1% reduced microvascular complications in T1DM and T2DM by 25%.

Retinopathy

Diabetic retinopathy is a potentially preventable cause of blindness. The prevalence is around 50% in T1DM and 33%
in T2DM; risk is strongly associated with diabetes duration. Diabetic retinopathy is caused by small vessel occlusion,
ischaemia leading to new vessel formation, and capillary leakage and fibrosis (Figure 50.1e). A number of changes occur:

• Background retinopathy: micro-aneurysms (‘dots’), small intraretinal haemorrhages (‘blots’) and lipid exudates forming around a leaking blood vessel (hard exudates).
• Maculopathy: background retinopathy evident within one disc diameter of the macula.
• Pre-proliferative retinopathy: cotton wool spots, clusters of vessels within the retina that may be early new vessels (intraretinal microvascular abnormalities) and venous changes such as beading and loops.
• Proliferative retinopathy: new vessel formation at the disc or elsewhere.

Screening

Patients are usually asymptomatic until significant damage has occurred; screening is thus important to allow early intervention. National retinal screening programmes record digital retinal images annually. These are graded, with reports returned to the diabetes team for management. Patients with potentially sight-threatening disease (pre-proliferative and/or proliferative retinopathy or maculopathy) are referred for pan-retinal photocoagulation (laser) treatment. In cases of intravitreal haemorrhage, vitrectomy may be needed. Good glycaemic and BP control are important in helping prevent development and progression.

Diabetic foot disease

Epidemiology

Around 25% of patients with diabetes will develop a foot ulcer during their lifetime. This imposes a significant financial burden on health services, with foot complications accounting for up to 20% of the NHS diabetes budget. Up to 85% of major amputations are preventable, leading to campaigns being introduced, such as ‘Putting Feet First’, to improve foot care.

Aetiology

Diabetic foot ulcers form as a result of DPN, peripheral vascular disease (PVD) or, more commonly, a combination of the
two. This leads to neuropathic, ischaemic or neuroischaemic ulceration, respectively. DPN affects sensory, motor and autonomic nerves. In the periphery, this results in a loss of protective sensation, leaving the patient exposed to heat, physical and chemical trauma without perception. Motor neuropathy can result in deformities that lead to abnormal pressure over bony prominences. Autonomic nerve damage causes a loss of sweating, dry skin and the appearance of
cracks or callus, which in turn can lead to neuropathic ulceration (Figure 51.5). PVD is associated with a reduced arterial supply, predisposing the patient to ischaemic ulceration. Even when the major arteries appear intact, small vessel dysfunction (microangiopathy) can be present, which impairs foot perfusion and delays ulcer healing.
Typically, up to half of diabetic foot ulcers are neuroischaemic, with both DPN and PVD present. Susceptibility to, and
progression of ulceration is increased by extrinsic factors, such a poor footwear or injury, and superimposed infection.

Clinical presentation

Patients with DPN describe numbness or painful ‘electric-shock’- type neuropathic pain, often worse at night. Some patients may be unaware of sensory loss and present with a neuropathic ulcer (Figure 51.5). Those with PVD can experience intermittent claudication, pain, pallor or cold extremities. Charcot’s foot occurs in severe DPN when an initial insult, such as minor trauma, causes a fracture, leading to progressive bony deformity and destruction (Figure 51.1). Patients can present with an acutely swollen, hot, red foot, which is painful in around one-third.

Management

Education

Patients should be educated regarding foot care, including advice on daily foot inspection, awareness of loss of sensation, looking for foot shape changes, keeping feet covered in well-fitting footwear, maintaining good blood glucose control and attending their annual foot review (Figure 51.2a).

Screening

An annual foot review should include a foot examination with shoes and socks removed to identify any risk factors for, or the presence of, diabetic foot disease. Sensation should be tested using a 10 g monofilament or 128 Hz tuning fork, foot pulses palpated (± using a hand-held Doppler ultrasound; Figure 51.3), feet inspected for any signs of deformity, callus, inflammation, infection, ulceration or gangrene as well as enquiring about past history of ulceration and pain and inspecting footwear (Figure 51.2b). Based on this assessment, a risk score can be calculated:

• Low: no risk factors present.
• Moderate: one risk factor present.
• High: previous ulcer or amputation, on renal replacement
  therapy (RRT) or more than one risk factor present.
• Active diabetic foot problem: ulceration, spreading infection, critical ischaemia, gangrene, suspicion of an acute Charcot arthropathy or an unexplained red, hot, swollen foot with or without pain.
  Those at low risk should be given advice and re-assessed annually, while those in moderate and higher risk groups should be referred to a foot protection service for ongoing advice, surveillance and assessment. In those with active foot disease, a rapid referral should be made to the hospital multidisciplinary foot team.

Assessment and management

The severity of an ulcer should be documented against a standardised system such as SINBAD (site, ischaemia,
neuropathy, bacterial infection, area and depth) or the University of Texas classification system (Table 51.1).

Infection

If infection is thought to be present, deep wound swabs or bony fragments (if bone is affected) should be sent for microbiological culture. Antibiotic choice is guided by local antimicrobial policy, the severity of infection and bone involvement (osteomyelitis). If osteomyelitis is suspected, a plain X-ray of the foot should be requested (Figure 51.4), although MRI is often needed to confirm the diagnosis. Prolonged antibiotics (for at least 6 weeks) are usually needed when osteomyelitis is present, and good glycaemic control is critical to aid wound healing.

Ischaemia

If there is evidence of ischaemia to the foot, a vascular surgeon should be involved to guide further imaging and revascularisation (angioplasty or bypass) as necessary.

Debridement

Local wound debridement can be undertaken by diabetes specialist podiatrists. In severe cases, patients need surgical
debridement, abscess drainage or, in unsalvageable wounds, amputation. Wounds can be treated with larval therapy, and a variety of dressings used to aid healing.

Off-loading

Ongoing advice to the patient to avoid weight-bearing, especially in neuropathic ulceration, is vital. If Charcot’s arthropathy is confirmed, the patient should be treated with a non-removable off-loading device to ensure immobilisation. In those who undergo amputation, post-surgical rehabilitation care is important.

Prognosis

Established diabetic foot disease carries a poor prognosis. In those with foot ulceration, around 40% will develop a second episode within 1 year, with 50% of patients dying within 5 years of presentation.

Diabetic ketoacidosis

Definition and epidemiology

Diabetic ketoacidosis (DKA) comprises the biochemical triad of hyperglycaemia (>11 mmol/L), ketonaemia (>3 mmol/L) and acidosis (pH <7.3 ± bicarbonate <15 mmol/L) (Figure 52.1a). It affects 0.5–0.8% of patients with T1DM annually. Around 25–30% children with newly diagnosed T1DM present in DKA.

Aetiology

DKA is characterised by a relative or absolute insulin deficiency of insulin, resulting in impairment of glucose utilisation in the peripheral tissues (Figure 52.1b). This leads to increased gluconeogenesis and glycogenolysis in the liver with consequent worsened hyperglycaemia. Simultaneous counter-regulatory hormone hypersecretion (including cortisol, glucagon and catecholamines) in tandem with insulin deficiency causes release of free fatty acids (FFAs) into the circulation as a result of lipolysis in adipose tissue. FFAs undergo oxidation in the liver to produce ketone bodies (β-hydroxybutyrate, acetoacetate and acetone) and subsequent ketonaemia. As ketone bodies are weakly acidic,
this causes increased plasma hydrogen ion concentrations and metabolic acidosis. Any state that causes a relative or absolute deficiency of insulin can lead to DKA, but infection is the most common precipitant.

Symptoms and signs

DKA usually develops rapidly, typically within 24 hours. Symptoms relate to hyperglycaemia and metabolic acidosis,
and include polyuria, polydipsia, weight loss, lethargy, vomiting, dehydration, abdominal pain and altered mental
state. Examination reveals dry mucus membranes, an odour of ketones, tachycardia, hypotension, Kussmaul breathing and focal signs of a precipitant, such as infection.

Investigations

Bedside meters can be used to measure both capillary glucose and ketones, while a venous blood sample will measure pH or bicarbonate. When blood ketone meters are not available, a urine dipstick can be performed, with urine ketones ++ or more being significant. An arterial blood sample is not usually required, as venous and arterial pH and bicarbonate correspond closely. An initial raised capillary glucose value should always be confirmed with laboratory measurement of plasma glucose from a venous blood sample. Further blood samples, including a FBC and renal
function, should be obtained and a septic screen performed if infection is suspected (blood and urine cultures, chest X-ray). An ECG can show evidence of tachycardia from dehydration, or arrhythmia resulting from electrolyte disturbance.

Management

The main aims of management are restoration of circulatory volume, clearance of ketones and correction of the electrolyte disturbance. Fluid should be replaced intravenously as crystalloid, with the initial aim of correcting any hypotension and replenishing the intravascular deficit. Fluids should be replaced cautiously in young adults, the elderly or those with evidence of cardiac or renal failure. Insulin is commenced as a fixed-rate intravenous infusion
(FRIII) based on the patient’s body weight. A rate of 0.1 units/ kg body weight/hour should be used, with an aim of reducing hyperglycaemia, suppressing ketosis and correcting any electrolyte disturbance. The FRIII is continued until DKA is fully resolved. If the patient is on a subcutaneous long-acting analogue or human insulin, this should be continued throughout treatment. When capillary glucose levels fall below 14 mmol/L, 10% glucose should be added to the fluid regimen. Careful monitoring of electrolytes is vital throughout treatment, as intravenous insulin may result in marked hypokalaemia. Electrolytes should be measured 4-hourly and potassium supplemented accordingly. Intravenous bicarbonate is not routinely recommended. Capillary glucose and ketones should be checked hourly until
there is complete resolution (defined as ketones <0.6 mmol/L and venous pH >7.3), which usually occurs within 24 hours of treatment commencing. When DKA has resolved and the patient is eating and drinking normally they should be converted to a regular subcutaneous insulin regimen. Involvement of the diabetes team, including the diabetes specialist nurse, is vital in educating the patient, particularly in relation to ‘sick day rules’ (Chapter 56) with a view to prevention of recurrence. For children, a different treatment algorithm is used.

Prognosis

The overall mortality in adults has fallen from around 8% to <1% over the past 20 years. However, a higher death rate is still apparent in the elderly and in those with co-morbidities. DKA remains the leading cause of mortality and morbidity in children with T1DM.

Hyperglycaemic hyperosmolar state

Definition and epidemiology

Hyperglycaemic hyperosmolar state (HHS) is a diabetes emergency characterised by the triad of (i) hypovolaemia; (ii)
marked hyperglycaemia (>30 mmol/L) without significant hyperketonaemia (<3 mmol/L) or acidosis (pH >7.3, bicarbonate 15 mmol/L); and (iii) osmolality >320 mosmol/kg (Figure 53.1a). It affects around 1 in 500 people with T2DM, with a mean age of 60 years at presentation.

Aetiology

HHS typically occurs in the elderly, but can occur in younger adults and teenagers, often as the initial presentation of T2DM. In contrast to DKA, which develops rapidly, HHS develops over many days. Prolonged hyperglycaemia from insulin resistance, or beta-cell failure and insulin deficiency from temporary glucose toxicity, results in an osmotic diuresis with renal sodium and potassium loss. This results in extracellular volume depletion and dehydration, with a raised serum osmolality. Ketosis/ ketonaemia does not typically occur in HHS, because some insulin is still present and hyperosmolality can inhibit lipolysis, although the reasons for this are not entirely clear. Typical precipitants for HHS include certain medications (e.g. thiazide diuretics), infection, surgery, MI/acute coronary syndrome, stroke and non-compliance with oral hypoglycaemics or insulin (Figure 53.1b).

Symptoms and signs

The clinical features reflect the hyperglycaemia, hyperosmolality and any underlying precipitant: polydipsia, polyuria, impaired cognitive function, tachycardia, hypotension, seizures and focal signs of thrombosis (Figure 53.1).

Investigations

Capillary blood glucose, plasma glucose and renal function should be measured. Serum osmolality should be calculated
using the formula (2[Na+] + glucose + urea). A venous blood gas (with lactate) should be taken to exclude significant acidosis, and blood ketones measured to exclude ketonaemia. Further investigations to establish the underlying cause of HHS should be performed, including a FBC and C-reactive protein, and a septic screen if infection is suspected (blood and urine cultures, chest X-ray). An ECG can show evidence of tachycardia from severe dehydration, or ischaemia in acute coronary syndrome. Troponin level can also be checked if cardiac ischaemia is suspected as a precipitant.

Management

The aims of treatment are to treat the underlying cause of the HHS, to gradually normalise the osmolality and glucose, and to replace the fluid and electrolyte losses. Increasing the circulating volume with rehydration leads to kidney reperfusion which can redress electrolyte abnormalities and excrete glucose. ‘Normal’ (0.9%) saline should be used as the fluid of choice. Fluid losses in HHS are thought to be around 100–220 mL/ kg. Intravenous fluids should aim to create a positive balance of 3–6 L within the first 12 hours, with the remaining fluid losses replaced over the next 12 hours. This should be tailored according to the patient, with caution needed in the elderly so as not to precipitate heart failure. Overly rapid correction can be harmful. ‘Half-normal’ (0.45%) saline is only recommended if serum osmolality does not improve despite an adequate positive fluid balance. Potassium shifts are less pronounced than in DKA
so potassium should be replaced only if needed (Figure 53.2c). Intravenous insulin is only recommended if the plasma glucose fails to fall following treatment with intravenous fluids alone or if significant ketonaemia is present. A rate of 0.05 units/kg/hour should then be commenced. An initial rise in serum sodium is usually expected as the plasma glucose starts to fall. Any underlying precipitant (e.g. infection) should be treated, and the patient treated with prophylactic anticoagulation (low molecular weight heparin) because the prevailing hyperosmolality and hypercoagulability lead to an increased risk of thromboembolic events. Assessment for complications (such as fluid overload or central pontine myelinolysis) should be carried out frequently. As all patients are at increased risk
of foot disease, daily foot checks should be undertaken and the heels protected. Resolution of HHS is much slower than DKA and is likely to be more than 24 hours. Good nutrition and early mobilisation are important. Intravenous insulin can be stopped when the patient is eating and drinking, and converted to subcutaneous insulin. After a period of stability, those with previously undiagnosed T2DM or previously well-controlled on oral hypoglycaemic agents can be considered for maintenance on oral therapy. Involvement of the diabetes specialist team and patient education to prevent recurrence and complications are important components of management.

Prognosis

Mortality, which is higher than in DKA at about 15–20%, is usually due to the underlying precipitating cause. Morbidity is also high, with complications including vascular disease (MI, stroke), seizures, central pontine myelinolysis and cerebral oedema.

Hypoglycaemia

Definition and epidemiology

Hypoglycaemia occurs when plasma glucose falls below 4 mmol/L. This can be defined as mild when self-treated, or
severe when third-party assistance is required. On average, patients with T1DM experience around two episodes of mild hypoglycaemia a week. Hypoglycaemia occurs in almost 8% of hospital admissions; the annual prevalence of severe hypoglycaemia is 30–40%.

Aetiology

In individuals without diabetes, the normal response to hypoglycaemia comprises reduced insulin secretion from the
pancreas and increased glucagon release. A number of counterregulatory hormones, including noradrenaline, cortisol and growth hormone, are also released. In patients with diabetes, these responses are reduced, especially with recurrent hypoglycaemia and with increased duration of disease.
Hypoglycaemia commonly occurs as a result of insulin therapy. This may be because of excess administration (e.g. dose
error), absorption problems (e.g. different site of administration), reduced clearance (in kidney disease) or decreased insulin requirement (e.g. during exercise). It can also occur with certain oral hypoglycaemic agents (notably sulphonylureas). Other factors unrelated to diabetes may need to be considered when the cause is not immediately clear (Chapter 34).

Symptoms and signs

The clinical features relate initially to the response of the autonomic nervous system to hypoglycaemia, followed by
that of the brain resulting from insufficient supply of glucose (neuroglycopaenia). Autonomic symptoms include sweating, feeling hot, anxiety, palpitations, shaking and paraesthesia. Neuroglycopaenic symptoms include difficulty speaking, poor concentration, poor coordination, drowsiness, fits and coma. Other symptoms include nausea, fatigue and hunger (Figure 54.1). As the counter-regulatory hormone and sympathetic neural response is impaired in diabetes, some of the autonomic symptoms and signs are absent or occur at a much lower level of plasma glucose, such that the patient is unaware of the hypoglycaemia. This is termed hypoglycaemia unawareness or impaired awareness of hypoglycaemia (IAH). This is more common in T1DM than T2DM.

Investigations

Hypoglycaemia must be recognised and treated quickly. A blood glucose meter should be used to confirm the capillary
glucose reading, where it is safe to do so (Figure 54.2). If the patient is compromised (e.g. having a fit or in a coma), rapid treatment should take priority. It may be appropriate to consider other investigations to aid in establishing a precipitant, such as performing an FBC and C-reactive protein if underlying infection or sepsis is suspected, or checking renal function if there is potential for renal impairment (leading to accumulation of insulin or sulphonylurea). Investigations for coeliac disease, hypoadrenalism and malignancy should also be considered in patients with new or recurrent hypoglycaemia and weight loss.

Management

Any suspected hypoglycaemia should be managed as an emergency, and treated immediately with a quick-acting
carbohydrate to return the blood glucose to the normal range. Short-acting carbohydrates include 150–200 mL pure fruit juice, 90–120 mL Lucozade or 4–5 Glucotabs. If the patient is uncooperative or unable to swallow, GluoGel can be squeezed into the mouth between the teeth and gums or, if this is ineffective, 1 mg intramuscular glucagon can be administered. Blood capillary glucose should be repeated after 10–15 minutes. If this is still <4 mmol/L, a further short-acting carbohydrate should be given. If the blood glucose is >4 mmol/L and the patient has recovered, a long-acting carbohydrate in the form of a snack or part of the next planned meal should be given. Suitable longer-acting carbohydrates include two biscuits, a slice of toast or 200–300 mL glass of milk.
If the blood glucose fails to rise above 4 mmol/L after three cycles of short-acting carbohydrate, or after 45 minutes, then 1 mg intramuscular glucagon should be administered. In a hospital, 10% dextrose intravenously can be commenced, usually at a rate of 100 mL/hour.
If the patient is unconscious or having a seizure, either intravenous 10% or 20% dextrose or 1 mg intramuscular
glucagon should be given immediately. However, glucagon will be less effective in patients with depleted liver glycogen stores, such as those with alcohol dependence or malnutrition.
In the hospital setting, a ‘hypo box’ should be available in all clinical areas, which should contain all the necessary equipment necessary to treat hypoglycaemia (Figure 54.3).
Once the acute hypoglycaemic episode has been treated, the underlying cause should be sought. The patient should be
educated to prevent further episodes, by involving the DSN and the diabetes team. Patients should be encouraged to monitor their blood glucose levels regularly and to note any patterns of recurrent hypoglycaemia in order to adjust insulin doses or diet accordingly. Education should be given to ensure a safe lifestyle, addressing issues such as exercise and driving. The next insulin dose due should not be omitted, although a dose reduction may be warranted depending on the underlying cause.

Prognosis

Most patients recover fully but permanent neurological sequelae (e.g. seizure, coma, hemiparesis) can result if severe
hypoglycaemia is left untreated for a long period of time.

Peri-operative management

The prevalence of diabetes in surgical patients is at least 10% and likely to increase. Patients with diabetes experience
higher morbidity and mortality (up to 50% higher) and longer lengths of stay than their non-surgical counterparts. Reasons for this include higher co-morbidity (including ischaemic heart disease, heart failure, respiratory disease, renal impairment), susceptibility to pressure sores and a greater risk of postoperative infections. In addition, inappropriate use or misuse of insulin and polypharmacy in the peri-operative period often puts patients with diabetes at risk.

Pathophysiological factors

The peri-operative period comprises both a catabolic state resulting from metabolic stress and a starvation period,
accompanied by increased catabolic hormone secretion and decreased anabolic hormone secretion, including insulin.
As the postoperative period is also associated with insulin resistance, the overall effect of surgery is one of relative insulin insufficiency.

Pre-operative management

Pre-operative assessment should be undertaken at several levels: by the GP in primary care, the surgeon in outpatients
and at the pre-operative assessment clinic (Figure 55.1). Three goals are important: (i) optimisation of glycaemic control; (ii) identification and optimisation of co-morbidites; and (iii) establishment of a diabetes plan for the pre-admission and perioperative periods (Figure 55.1).
An HbA1c >8.5% (69 mmol/mol) may be a suitable threshold to refer the patient to the diabetes specialist team for intensification of glycaemic control, assuming the timing of surgery allows. Similarly, patients with impaired awareness of hypoglycaemia (IAH) should be referred for diabetes specialist input. In the preoperative assessment clinic, the patient should be assessed for suitability for day surgery and plans made to ensure admission on the day of surgery. Patients should be prioritised to ‘early’ on the operating list to avoid prolonged starvation, unnecessary use of
intravenous insulin regimens and a longer inpatient stay.
Written information should be provided for the patient, particularly with regard to modifications needed to their usual diabetes medication on the day prior to and on the day of surgery. On the day of surgery, for patients with good glycaemic control (HbA1c <8.5%) and with a short starvation period planned, it is recommended that: (i) no change in dose is needed for patients receiving once daily insulin; (ii) the usual morning dose is halved for those on twice daily insulin, leaving the evening dose unchanged; (iii) those on twice daily injections of separate short-acting and intermediate-acting insulins should have the total dose of both morning doses calculated and half given as
intermediate-acting insulin in the morning; and (iv) those on a basal bolus regimen should omit the breakfast and lunchtime doses for morning surgery and only the lunchtime dose for an afternoon planned procedure (Table 55.1).
Metformin and pioglitazone can be given as normal on the day of surgery. GLP-1 receptor analogues and DDP-4 inhibitors should be omitted on the day of surgery and recommenced only when the patient is eating and drinking normally. If a patient is on a once daily sulphonylurea, this should be omitted on the day of surgery; if twice daily, only the evening dose needs to be omitted if afternoon surgery is undertaken (Table 55.2).
If the patient has poor glycaemic control or a longer period of starvation is planned (missing more than one meal) then a variable-rate intravenous insulin infusion (VRIII) is required. Long-acting insulin analogues can be continued alongside the VRIII. Careful fluid management and monitoring of capillary blood glucose (CBG) are needed during this period. For patients on CSII pump therapy, if a short starvation period is planned (omitting only one meal), then the pump can be continued with normal basal rates and with close monitoring of CBG. If more than one meal is to be missed, the pump should be removed and VRIII commenced. In cases of emergency surgery, blood glucose levels should be monitored; if they rise above 10 mmol/L a VRIII should be started.

Intra-operative management

During surgery, blood glucose concentrations should be maintained at 6–10 mmol/L. CBG should be monitored at
least hourly pre-theatre, during induction, throughout surgery and in the recovery phase. High blood glucose should be
corrected using additional subcutaneous insulin or a VRIII when required. Fluids should be prescribed appropriately
throughout, particularly to maintain optimal cardiac and renal function. Avoidance of hypotension and decreased
skin perfusion is particularly important for those with a CSII pump, as this will affect absorption of insulin during surgery. If present, hypoglycaemia should be treated by commencement of an intravenous glucose infusion. Consideration should also be given to the type of anaesthetic (general, regional or local), the use of adequate analgesia and anti-emetics so as to promote early eating and drinking, and an early return to usual diabetes therapy.

Postoperative management

Blood glucose levels should also be maintained at 6–10 mmol/L during the postoperative period, with the goal of recommencing the usual diabetes treatment as soon as possible. A high dependency setting may be appropriate for patients at higher risk of postoperative complications. Careful attention should be paid to monitoring fluid balance, electrolytes, foot care and preventing infection. Appropriate pain relief and nausea management is also key. Self-management of diabetes should be encouraged where possible and every surgical ward should have input available from the diabetes specialist team. Discharge should be planned early; poor glucose control should not hinder discharge if either the patient or their carer are able to self-manage their diabetes appropriately. The patient should be educated about postoperative factors that can affect glucose control (e.g. pain, infection, nutritional intake) and ‘sick day rules’ should be discussed.

Management of acute illness

A range of acute illnesses can affect patients with diabetes, with infections such as the common cold, influenza, gastroenteritis, urinary tract infections, chest infections and abscesses being the most common. Patients need to know how to manage their diabetes during such illnesses, to prevent hyperglycaemia, dehydration or the development of emergencies such as DKA or HHS (Chapters 52 and 53). Education on ‘sick day rules’ is therefore important.

Aims of management

The main goals are controlling blood glucose levels, ensuring adequate calorific intake and hydration, testing for ketones and recognising when medical help is required (Figure 56.1).

Monitoring

During an infection, blood glucose levels can rise, even in the absence of food, sometimes resulting in significant
hyperglycaemia. Those with access to blood glucose monitoring should check their capillary blood glucose levels at least four times a day (at meal times), even if they are not eating. If they have no access to blood glucose monitoring, they should be advised about the symptoms of hyperglycaemia.
Ketone monitoring should be advised in any patient with T1DM and a blood glucose >13 mmol/L. Blood ketones should
be checked by a health professional for anyone with diabetes who is vomiting. If urinary or blood ketones are present, extra doses of pre-mixed or short-acting insulin should be given, depending on the usual total daily dose of insulin. Blood glucose and ketone levels should then be monitored 2- to 4-hourly and corrected as necessary. If blood glucose or ketone levels are not controlled, or vomiting persists, urgent medical advice should be sought.

Drug management

The following advice should be heeded for patients on oral hypoglycaemics:

• Metformin should be continued when the glucose level is normal or high, unless the patient becomes severely unwell with vomiting or dehydration, or requires confinement to bed or hospitalisation, in which case it should be stopped. It should particularly be stopped in the presence of acute renal failure or hypoxia.
• Sulphonylureas should be continued when the glucose level is normal or high, unless patients are unable to eat or drink, with risk of hypoglycaemia, when sulphonylureas may need to be reduced or temporarily withheld.
• Thiazolidinediones should be continued when the glucose level is normal or high but should be stopped if the patient becomes acutely breathless, with signs of fluid overload. All of the above therapies are contra-indicated in patients with DKA.
• DPP-4 inhibitors can be continued when the glucose level is normal or high, but patients should seek medical advice in the presence of acute vomiting or abdominal pain, because this can suggest acute pancreatitis.
• GLP-1 receptor agonists should be continued when the glucose level is normal or high, but medical advice should be sought if any symptoms of dehydration, severe abdominal pain or vomiting develop.
• SGLT-2 inhibitors should be stopped during periods of acute illness and dehydration given the risk of further volume depletion.
  For patients taking insulin, doses may need to be increased in the presence of hyperglycaemia. The dose increase depends on whether the patient has T1DM or T2DM and taking a premixed or short-acting insulin. If blood glucose levels are lower than usual or there is hypoglycaemia, doses should be reduced or stopped. As the acute illness subsides, insulin doses can then be gradually tapered back to normal.

Calorie and fluid intake

An adequate intake of carbohydrate should be encouraged during acute illness. If the patient is unable to eat their usual meals, they should maintain carbohydrate intake by eating or drinking sugary or starchy food (Table 56.1). They should aim to take two to three servings of carbohydrate four to five times a day. Adequate oral hydration should be maintained by drinking at least 100 mL/hour of sugar-free fluid, or 2.5 L/day in total. In the presence of persistent vomiting, there is a risk of dehydration, DKA or HHS, hence further medical advice should be sought.

Special groups for consideration

The following groups require special consideration.

• Pregnant women: medical advice should be sought early if patients feel unwell, even when they have normal or mildly elevated glucose levels. They should be provided with an emergency contact number (obstetrics or diabetes team).
• End of life care: the main goal should be symptomatic control of hyperglycaemia, to ensure the patient is as comfortable as possible. Efforts should be made to reduce symptoms, particularly of thirst and dehydration, while trying to avoid the development of DKA or HHS. Management should be based on each individual patient
  and their needs.
• Insulin pumps: patients with insulin pumps should be advised that if the pump fails, DKA can develop very rapidly. If glucose levels rise acutely, patients should check for blood or urine ketones, check the pump to ensure it is working properly, check the tubing to ensure it is connected and not blocked or kinked, and ensure that the cannula is fixed in the right place. Patients should have access to a short-acting insulin pen for use in the event of pump failure, and an emergency contact number to seek medical help.
• Other medications taken during acute illness can affect glucose levels. For example, a course of steroid therapy in an exacerbation of chronic obstructive pulmonary disease can cause hyperglycaemia. Individual advice should be given on how best to control glucose in such circumstances. ACE inhibitors or angiotensin-receptor blockers should be temporarily withheld in any dehydrating illness in order to reduce the risk of acute kidney injury.

Insulin infusions

Patients with diabetes occasionally require an intravenous insulin infusion such as during DKA or HHS. This can
be infused either: (i) at a variable hourly rate – known as variable-rate intravenous insulin infusion (VRIII) (‘sliding
scale’), or (ii) at a fixed hourly rate – fixed-rate intravenous insulin infusion (FRIII).

Variable-rate intravenous insulin infusion

Indications and aim

VRIII is required when a patient with diabetes is intolerant of oral fluids, is nil by mouth with prolonged starvation or has decompensated diabetes. The aim is to achieve and maintain normoglycaemia (range 6–10 mmol/L).

Principles

Each unit should develop its own protocol, although national Joint British Diabetes Societies guidelines will encourage more standardised practice. Some patients require deviation from the standard protocol, especially if they are overweight and more insulin-resistant. If a patient is already on a long-acting insulin analogue (e.g. glargine or detemir), this should be continued alongside the VRIII. The initial infusion rate should be based on a bedside capillary blood glucose (CBG), repeated hourly. If glucose remains >12 mmol/L for two consecutive readings or is not falling by >3 mmol/L/hour then the infusion rate should be increased. For some patients, CBG of 4–6 mmol/L may be too low
(e.g. following acute coronary syndrome). In such circumstances, the VRIII rate can either be decreased or an increased substrate can be added, such as 10% dextrose (Table 57.1). If the CBG falls below 4 mmol/L, the infusion should be stopped and the patient treated for hypoglycaemia. Once the CBG is >4.0 mmol/L, the infusion should be restarted within 20 minutes to prevent rebound hyperglycaemia or ketosis.

Administration

Fifty units of soluble insulin in 50 mL 0.9% normal saline (1 unit/ mL) should be placed in a 50-mL syringe and run through an automated syringe driver pump at an initial rate based on the bedside CBG (Figure 57.1). A substrate fluid should be selected to run with the VRIII; this will depend on local protocols. Some advocate using 0.9% saline and switching to 5% dextrose when the CBG falls below 12 mmol/L. However, this can increase the risk of hypoglycaemia, hence a substrate such as 0.45% saline, 5% dextrose or 0.3% KCl is often advisable with careful monitoring of
electrolytes to avoid hypokalaemia. The substrate fluid infusion rate depends on the circulatory volume and clinical situation.

Switching back to usual insulin regimen

Once the patient is eating and drinking adequately, their usual subcutaneous insulin regimen can be recommenced. The patient should be given their usual subcutaneous insulin with their meal and the VRIII stopped 30 minutes later. If the patient is on oral hypoglycaemics, careful monitoring of CBG should be continued and a short period of subcutaneous insulin may be required. Short-term reduction in sulphonylurea dose may be needed when food intake is reduced.

Patients starting insulin for the first time

The estimated total daily dose (TDD) of insulin is based on various factors, including the patient’s weight, age and estimated insulin sensitivity. TDD can be calculated by dividing the total amount of insulin required over the past 6 hours on the VRIII by 6. This will give the average hourly insulin dose, which should then be multiplied by a conservative 20 to give the TDD. For a basal bolus regimen, 50% of the TDD should be given as a once-daily long-acting insulin and the remainder divided equally into boluses of short-acting insulin with each meal. For a twice daily pre-mixed insulin, two-thirds of the TDD should be given in the morning and one-third with the evening meal.

Fixed-rate intravenous insulin infusion

Indications and aim

FRIII is indicated in cases of DKA or imminent DKA. The aim is to treat DKA by reducing ketogenesis and clearing ketones, normalising hyperglycaemia and restoring electrolyte balance. This is achieved when blood ketones are <0.6 mmol/L and venous pH >7.3.

Principles

The FRIII is based upon the patient’s body weight, and should be commenced at 0.1 unit/kg/hour (Table 57.2). This remains fixed throughout treatment until resolution. However, this rate should be increased by 1 unit/hour if blood ketones fail to fall by >3 mmol/L/hour, if the CBG fails to fall by >3 mmol/L/hour or venous bicarbonate fails to rise by >3 mmol/L/hour. Bedside CBG and blood ketone monitoring should be performed hourly to ensure these targets are being met. If the patient takes a long-acting insulin analogue (e.g. glargine, detemir), this should be continued. If the patient has a new diagnosis of T1DM, they should be commenced on a long-acting insulin analogue or NPH insulin at a dosage of 0.25 units/kg/day alongside the FRIII to prevent rebound ketosis when the FRIII is stopped.

Administration

As with VRIII, 50 units soluble human insulin (Actrapid or Humulin S) should be made up with 50 mL 0.9% saline in an
infusion pump. This should be infused at a rate of 0.1 units/ kg/hour of insulin, with the CBG and blood ketones measured hourly, as described, and the rate only adjusted if needed. The fluid substrate of choice to run alongside the FRIII is 0.9% saline and the rate should be given according to the management of DKA. When the CBG falls below 14 mmol/L, 10% glucose should be infused concurrently with 0.9% saline to prevent hypoglycaemia and enable the FRIII to continue safely. Potassium levels should be monitored carefully and corrected as for DKA management.

Switching back to usual insulin regimen

Once DKA has resolved and the patient is eating and drinking normally, they can be recommenced on their usual insulin regimen at the next meal. Their usual insulin dose should be given at the meal and the FRIII stopped 30–60 minutes later.

Patients starting insulin for the first time

The TDD can be calculated by multiplying the patient’s weight in kilograms by 0.5–0.75 units. As above, for a basal bolus regimen, 50% of the TDD should be given as a long-acting insulin and the remainder divided into three mealtime bolus doses. For a twice daily pre-mixed insulin, two-thirds of the TDD should be administered in the morning and the remaining one-third with the evening meal.

Pregnancy and diabetes

Epidemiology

Diabetes in pregnancy can pre-exist in patients with T1DM or T2DM, or develop as gestational diabetes mellitus (GDM),
which affects about 3–4% of pregnant women.

Pathophysiology

Pregnancy is associated with increased insulin resistance from the second trimester onwards, accompanied by an increase in insulin secretion (up to 250%). These changes are influenced by maternal and placental factors, such as production of placental growth hormone which increases peripheral insulin resistance. These physiological changes help to maintain maternal euglycaemia but can result in GDM if insulin secretion fails to meet the increased demand. Risk factors for GDM are as for T2DM, and include obesity, a first-degree relative with T2DM, previous GDM and ethnicity (South Asian, African- Caribbean).
As a result of increased insulin resistance, insulin requirements increase from the second trimester onwards in patients with pre-existing diabetes. In contrast, maternal insulin requirements may be less in the first trimester, which places the mother at increased risk of hypoglycaemia.

Complications

Diabetes in pregnancy is associated with an increased incidence of stillbirth and congenital malformations, which occur in about 4% of births. These include sacral agenesis, neural tube defects, cardiac and renal anomalies (Figure 58.1a). Hyperglycaemia in the second and third trimesters can lead to accelerated fetal growth, resulting in macrosomia. This increases the risk of fetal malpresentation and shoulder dystocia. In the neonate, there is an increased risk of hypoglycaemia, respiratory distress syndrome and polycythaemia leading to jaundice.

Pre-conception care

Good pre-conception care is critical in reducing the potentially teratogenic effects of hyperglycaemia. Women with diabetes who are planning to become pregnant should aim for an HbA1c <48 mmol/mol (6.5%) whereas those with an HbA1c of 86 mmol/mol (10%) should avoid pregnancy until glycaemic control is improved (Figure 58.1). Overweight women with diabetes should be encouraged to lose weight before becoming pregnant. Folic acid (5 mg/day) is advised in all patients until 12 weeks’ gestation to reduce the risk of neural tube defect. Potentially teratogenic drugs, such as statins and ACE inhibitors, should be discontinued before pregnancy or as soon as pregnancy is confirmed.

Screening for GDM

Women at risk of GDM should be offered screening with a 75 g 2-hour OGTT at 24–28 weeks’ gestation (Figure 58.1c). These risks include:

• BMI >30 kg/m²
• Previous macrosomia (baby ≥4.5 kg)
• Previous GDM
• Family history of diabetes in a first-degree relative
• Higher risk ethnicity.
  If the mother has had GDM in a previous pregnancy, an OGTT should be offered in the first or second trimester, and
  repeated at 24–28 weeks if this is normal. In the UK, GDM is diagnosed if:
• Fasting plasma glucose ≥5.6 mmol/L, or
• 2-hour glucose ≥7.8 mmol/L.

Antenatal care

Women should be monitored in a joint diabetes–obstetric antenatal clinic regularly throughout pregnancy. Good blood
glucose control is vital, aiming to avoid hypoglycaemia and targeting a fasting glucose of <5.3 mmol/L, 1-hour post-meal <7.8 mmol/L or 2-hour post-meal <6.4 mmol/L (Figure 58.1b). HbA1c can be measured in patients with pre-existing diabetes in early pregnancy, or during the second or third trimesters, to establish the level of risk for that pregnancy. However, in later trimesters it should not be relied upon as a measure of glycaemic control because of decreased red blood cell lifespan and increased erythropoietin production.
Lifestyle advice (diet, exercise) should be provided but metformin should be commenced in women with GDM who fail to meet glycaemic targets, adding insulin therapy if glycaemic control continues to be suboptimal. If there are complications, such as macrosomia or polyhydramnios, immediate treatment with insulin may be necessary. Glibenclamide can also be considered for women who fail to tolerate or achieve targets with metformin, or who
refuse insulin therapy. Insulin pump (CSII) therapy can be offered to insulin-treated patients who fail to achieve glycaemic targets without disabling hypoglycaemia. Women taking insulin should be advised about the risks of hypoglycaemia, particularly during the first trimester, and its management. Advice should also be provided
on managing intercurrent illness, including the importance of monitoring capillary ketones in patients with T1DM.
An increased frequency of retinal screening is required as retinopathy can worsen during the antenatal period. Digital
screening should be offered early in pregnancy, at 16–20 weeks if retinopathy is present, and again at 28 weeks. An assessment of renal function is required in early pregnancy for women with pre-existing diabetes. A fetal ultrasound scan to screen for anomalies should be performed at 20 weeks and periodically thereafter to assess fetal growth and amniotic fluid volume.

Peri- and postnatal care

Induction of labour or caesarean section should be considered for patients with T1DM or T2DM from 37 weeks to reduce the incidence of stillbirth. During labour, blood glucose should be maintained at 4–7 mmol/L, using an intravenous insulin infusion if necessary. After delivery, the baby should be monitored for hypoglycaemia, and early breastfeeding encouraged at frequent intervals until the baby’s pre-feed glucose levels are >2.0 mmol/L.
Women who breastfeed on insulin should be advised about the increased risk of hypoglycaemia, and encouraged to take a snack before or during breastfeeding.
In women with GDM, all treatment can be discontinued postdelivery. In those with pre-existing diabetes, insulin requirements fall rapidly to pre-pregnancy levels after delivery. Women with T2DM can continue or restart metformin or glibenclamide while breastfeeding but should avoid other oral hypoglycaemics.

Postnatal diabetes screening

Women with GDM should be checked to ensure they do not have hyperglycaemia before discharge from hospital. A fasting plasma glucose (at 6–13 weeks post-delivery) or an HbA1c (at 13 weeks) should be measured to screen for ongoing diabetes. Patients should then be screened annually for diabetes as up to half will develop T2DM within 10 years.

Genetics of diabetes

Type 1 diabetes

1–2% if the individual’s mother has T1DM, 3–5% if the father has T1DM, with concordance of up to 35% in monozygotic twins.
Genes in the MHC/HLA glycoprotein system influence susceptibility to T1DM. HLA class II molecules bind foreign
peptides and present them to T-helper lymphocytes. The HLA class II DR and DQ loci on chromosome 6, encoded by DRB and DQB genes, are specifically associated with disease susceptibility (Figure 59.1). HLA-DR-3-DQ2/DR-4-DQ8 class II HLA antigens are found in over 95% of Europeans with T1DM. Conversely, some HLA haplotypes, including HLA DQ-5 and DQ-6, protect against T1DM.
Non-HLA loci linked to T1DM susceptibility include the insulin gene on chromosome 11, the CTLA4 gene (encoding
T-cell surface receptors), PTPN22 and PTPN2 genes (encoding T-lymphocyte tyrosine phosphatases) and interleukin 2 gene.

Type 2 diabetes

Heritability is thought to be significant in T2DM, accounting for 25–70% of disease susceptibility, with a concordance rate of up to 100% in monozygotic twins.
T2DM is a polygenic disorder with a large number of genes thought to be involved (Figure 59.1a). Many of these influence insulin secretion and action as well as hepatic and peripheral insulin resistance. TCF7L2 on chromosome 10q, encoding a transcription factor, is the susceptibility locus that seems to have the largest effect but still only accounts for up to 20% of the inherited risk.

Maturity onset diabetes of the young

MODY is inherited as an autosomal dominant condition and accounts for up to 2% of diabetes cases. Six genetic loci have been identified that account for almost 90% MODY cases in the UK but further genes remain undiscovered.
The three key features of MODY are (Figure 59.1c):

• 1 Diabetes usually diagnosed below age 25
• 2 Family history in each generation
• 3 Diabetes that does not always require insulin but can be treated with diet alone or oral medication.

The six main types currently recognised are:

• 1 HNF1A-MODY (MODY3): hepatic nuclear factor 1 alpha(HNF1A) mutations, on chromosome 12, account for up to 70% of MODY cases. It affects insulin secretion, and can lead to severe hyperglycaemia and microvascular complications. Patients often respond well to sulphonylureas, but insulin may eventually be required to maintain euglycaemia.
• 2 GCK-MODY (MODY2): this accounts for 30–70% of MODY cases, and affects the glucokinase (GCK) gene located on chromosome Glucokinase acts as a ‘glucose sensor’ for the β-cell. Mutations in this gene result in a ‘re-set’ of glucose-regulated insulin secretion to a higher threshold. Although fasting blood glucose levels are therefore higher than normal (usually 5.5– 8.0 mmol/L), the condition is not associated with microvascular complications and can be managed with dietary modification alone.
• 3 HNF1b-MODY (MODY5): a familial cystic kidney syndrome associated with mutations in the hepatocyte nuclear factor-1β gene, also known as RCAD (renal cysts and diabetes). It accounts for 5–10% of MODY cases, and patients present with familial renal cystic disease, early onset diabetes, genital tract malformations, hyperuricaemia and early onset gout.
• 4 HNF4A-MODY (MODY1): mutations in the hepatic nuclear factor 4 alpha (HNF4A) gene affect insulin secretion. This accounts for 5–10% of MODY cases and is initially sensitive to sulphonylureas but insulin therapy can be needed later. Affected neonates can be macrosomic and hypoglycaemic.
• 5 IPF1-MODY (MODY4): resulting from mutations in the insulin promoter factor 1 gene.
• 6 NEUROD1-MODY (MODY6): resulting from mutations in the neurogenic differentiation-1/β2 gene.
  Further information on MODY can be found at diabetesgenes.org.

Mitochondrial diabetes

Point mutations in mitochondrial DNA can lead to diabetes. The most common mutation occurs at position 3243 in the tRNA leucine gene, leading to an A- to -G transition (Figure 59.1b). This form of diabetes is maternally inherited and commonly associated with sensorineural deafness.

Neonatal diabetes

Diabetes diagnosed below 6 months of age is termed neonatal diabetes. There are two main types: transient or permanent. Transient neonatal diabetes accounts for 50–60% of cases and usually remits within 3 months, but can recur later on in life, often in teenagers. Permanent neonatal diabetes occurs because of mutations in the KCNJ11 and ABCC8 genes which encode subunits of the ATP-sensitive potassium channel. Patients can usually be treated with high-dose sulphonylureas. Some cases of permanent neonatal diabetes can be caused by INS gene mutations, which are treated with either insulin sensitisers or insulin.

Monogenic disorders of insulin resistance

A number of mutations cause severe insulin resistance. Patients typically present at a young age with acanthosis
nigricans, hypertension and dyslipidaemia. Lipodystrophies are characterised by insulin resistance with abnormal fat distribution, and are more commonly diagnosed in females. They occur as a result of mutations in the LMNA and PPARG genes for example. Childhood syndromes attributable to mutations in the insulin receptor gene and characterised by marked insulin resistance and hyperinsulinaemia are Rabson–Mendenhall syndrome (teeth and nail abnormalities, pineal gland hyperplasia) and leprechaunism (usually fatal in infancy).
The geneticist, genetic diabetes nurse and diabetologist are important in counselling, family screening and disease
management in suspected or confirmed monogenic disorders of diabetes.

The multidisciplinary team

Diabetes is a complex chronic disease that requires input at diagnosis and throughout life from an effective MDT. This
team should include a number of core members, each with their own set of knowledge and skills, with input from other services and specialities in the community and hospital as needs arise (Figure 60.1). Core members of the MDT typically include the physician, DSN, GP, dietitian, podiatrist, clinical psychologist and community nurse.
There are several aspects of diabetes care and delivery that illustrate the importance of the MDT approach.

Diagnosis and surveillance

Patients with newly diagnosed diabetes require education on their disease, its daily management (including lifestyle changes, driving, travel, work, glucose monitoring and sick day rules) and the importance of self-care. DSNs often provide much of this input at diagnosis and offer a consistent point of contact for the patient.

Structured education

In the UK, established structured education programmes ensure that patients with diabetes are educated in a uniform and high-quality manner. The Dose Adjustment For Normal Eating (DAFNE) course is an educational programme for patients with T1DM that provides patients with the skills necessary to estimate the carbohydrate in each meal and to inject the appropriate dose of insulin (‘carbohydrate counting’). Similarly, the DESMOND (Diabetes Education and Self-Management for Ongoing and Newly Diagnosed) course offers structured input for patients with T2DM, providing group education and a resource to help manage diabetes-related problems. X-PERT is a programme that runs in three different formats to cater for those with different types of diabetes and those at high risk of diabetes. These education programmes are typically delivered by locally trained dietitians and DSNs. Although the provision and delivery of
such programmes varies across the UK, all patients with diabetes should have access to these educational resources.

Life stages

Diabetes provides different challenges at different stages of life. Management priorities thus need to change accordingly.

Paediatrics and young adults

Children with diabetes require specialist input from the paediatric DSN, to include education of parents and teachers. Adolescence brings the challenges of independent living and self-management, as well as lifestyle changes such as education, work, driving, alcohol and sexual development, all requiring education and support from the MDT. Psychological issues often manifest for the first time at this age, and input from psychology can be required to provide
the necessary support. There are various models to support the transition between adolescence and adult care. It is vital that maximal support is provided during this period of change to prevent young adult patients from being ‘lost’ from the system.

Pregnancy

Pregnancy in diabetes requires organised MDT input, beginning with pre-conception planning. Pregnancy itself demands close monitoring of the patient by the diabetologist, obstetrician, DSN and midwife, often provided through joint antenatal clinics.

The older patient

Ageing can bring separate challenges, because of increasing co-morbidities, polypharmacy, risk of hypoglycaemia, cognitive decline and reduced ability to self-manage disease. Community district nurses, care home workers and GPs have key roles in managing these challenges.

Disease stages

As diabetes progresses, macro- and microvascular complications can develop. Screening for microvascular complications from an early stage involves several members of the MDT, including podiatrists (assessing and advising on foot care) and the retinopathy screening service (with input from ophthalmologists when management of established retinopathy is needed). Renal involvement can lead to established nephropathy and deteriorating renal function, hence early input and ongoing care is essential, particularly in patients requiring renal replacement
therapy.
Peripheral neuropathy and vascular disease can lead to a number of complications. Input from the foot MDT is important in managing such patients, including the physician (to manage glucose control, blood pressure and neuropathic pain), podiatrist, vascular surgeon and orthopaedic surgeon. This service should be readily accessible to patients based in the community or in secondary care.
Patients with T1DM who are eligible for islet cell or whole pancreas transplantation require assessment by regional MDTs, who will usually comprise the diabetologist with a special interest, DSN, transplant surgeon and radiologist.

Setting

The changing face of diabetes in the UK has meant that a large proportion of patients are increasingly managed in the
community. Various models of care exist, but there is a common emphasis on care being delivered away from the hospital setting wherever possible. Some regions of the UK employ a specific community diabetologist, with clinics held in the community often jointly with GPs and practice nurses, while others employ GPs with a special interest in diabetes. Complex cases can often be discussed in community MDTs without the need for hospital review.
In hospitals, at any one time 15–20% of inpatients have diabetes. A structured MDT approach in identifying, empowering and managing these patients is therefore important in order to facilitate good glycaemic control and timely discharge. This is an area of ongoing interest, with a number of guidelines having emerged in recent years to promote good inpatient care.

Lipid disorders

Lipid disorders are a group of disorders characterised by an excess of cholesterol, triglycerides and/or lipoproteins in the blood. They can be subdivided into primary (hereditary, with a genetic cause) or secondary (acquired).

Primary hyperlipidaemias

Primary hyperlipidaemias are traditionally classified according to the Fredrickson classification, divided into types I–V depending on the lipoprotein pattern (Table 61.1).

Type I hyperlipidaemia

This is caused by lipoprotein lipase deficiency or apolipoprotein C2 deficiency and results in elevated chylomicrons (triglyceride-rich lipoproteins that transport fatty acids from the gastrointestinal tract to the liver). Serum cholesterol is normal. It is inherited in an autosomal recessive manner and patients present in childhood with eruptive skin xanthomata and abdominal pain resulting from acute pancreatitis. Other complications include
retinal vein occlusion and lipaemia retinalis. Treatment is usually with dietary measures.

Type II hyperlipidaemia

This is subdivided into types IIa and IIb. Type IIa is more commonly known as familial hypercholesterolaemia (FH).
This is an autosomal dominant disorder affecting 1 in 500 of the population. Mutations in the low density lipoprotein
(LDL) receptor account for almost 90% of cases, resulting in reduced LDL uptake from the circulation. Less commonly, FH is caused by mutations in the PCSK9 (1%) or apoplipoprotein B-100 (3–4%) genes. Elevated circulating LDL puts patients at significantly increased risk of premature coronary heart disease (CHD), with more than half of heterozygotes dying of CHD before the age of 60 years if left untreated. Clinical stigmata include tendon xanthomata (which are pathognomonic of FH), corneal arcus and xanthelasmic deposits around the eyes (Figure 61.1). Diagnosis is based on the Simon Broome criteria: total cholesterol >7.8 mmol/L or LDL >4.9 mmol/L plus tendon xanthomata in the patient or a first- or second-degree relative, or DNA-based evidence of an FH mutation. First-line treatment is statin therapy in FH, aiming to reduce baseline LDL cholesterol by >50%. PCSK9 inhibitors are a new drug class which may be effective. Plasmapheresis can also be needed to achieve target levels. Family screening is an important component of management.
Type IIb, otherwise known as familial combined hyperlipidaemia, affects roughly 1 in 125 of the population and
patients usually present later than those with FH. It can manifest with premature CHD in the patient or family members. The biochemical picture includes raised triglycerides (contained in very low density lipoproteins, VLDL) as well as cholesterol.

Type III hyperlipidaemia

Also known as familial dysbetalipoproteinaemia or broad beta disease, this is an uncommon autosomal recessive disorder which affects 1 in 10 000 people. A defect in apolipoprotein E (ApoE) synthesis leads to raised intermediate density lipoproteins and chylomicron remnants, which are usually cleared from the circulation by the ApoE receptor. The lipid pattern is often a mixed hyperlipidaemia with total cholesterol >5 mmol/L and triglycerides >5 mmol/L. Clinical features include tuberous xanthomata found over the elbows and knees, palmar xanthomas and an increased risk of premature CHD. Treatment is with diet, fibrate drugs and/or fish oils.

Type IV hyperlipidaemia

Familial hypertriglyceridaemia is an autosomal dominant disease which affects <1% of the population. It is characterised by increased VLDL production and decreased elimination, resulting in a high triglyceride level. Eruptive xanthomata and acute pancreatitis can occur as a consequence. Alcohol and certain drugs (e.g. thiazide diuretics, glucocorticoids) can worsen the dyslipidaemia, hence a low fat, no alcohol diet is recommended. Fibrates, nicotinic acid and statins can help in reducing triglyceride levels.

Type V hyperlipidaemia

This is very similar to type I hyperlipidaemia but is also characterised by elevated VLDL in addition to chylomicron
levels. A mixed lipid disturbance is usually seen, with elevated total cholesterol, triglycerides and LDL, often accompanied by low high density lipoprotein (HDL) cholesterol. It is commonly associated with glucose intolerance, diabetes and obesity, and usually responds well to statin or fibrate therapy.

Other rare primary hyperlipidaemias

These include hyperalphalipoproteinaemia (mildly elevated HDL and total cholesterol), abetalipoproteinemia (low total
cholesterol associated with fat malabsorption and spinocerebellar degeneration) and familial hypobetalipoproteinemia (low total cholesterol, organomegaly, neurological changes and acanthocytic red blood cells).

Secondary hyperlipidaemias

10–20% of dyslipidaemias in adults. The dyslipidaemic pattern is either mixed or an increase in triglycerides or cholesterol alone.
Secondary causes include:

• Diet-induced
• Obesity
• Diabetes mellitus: usually raised triglycerides, low HDL and raised LDL cholesterol
• Drugs: glucocorticoids, thiazide diuretics, beta-blockers, oestrogens and protease inhibitors
• Alcohol: excessive consumption can result in hypertriglyceridaemia
• Chronic kidney disease: low HDL cholesterol and high triglycerides. Nephrotic syndrome can lead to hypercholesterolaemia
• Hypothyroidism: can result in hypercholesterolaemia.
  These acquired dyslipidaemias are treated by managing the underlying cause, which includes stopping any offending drug. In persistent cases, lipid-lowering drug therapies may be needed.

Appetite and weight

Appetite and weight are regulated by the CNS and gut (Figure62.1).

Central regulation

The hypothalamus is the key CNS centre involved in appetite regulation. Neuronal inputs from the brain and humoral factors from the gut regulate short-term appetite. Integration of these signals occurs in the arcuate nucleus of the hypothalamus. Two groups of appetite-regulating neurons are located here: (i) the pro-opiomelanocortin (POMC) and cocaine- and amphetamine-regulated transcript (CART) appetite-inhibiting neurons, and (ii) the neuropeptide Y (NPY) and agouti-related peptide (AgRP) appetite-stimulating neurons. Incoming signals change the relative activity of these two systems, affecting the release of neuropeptides that affect appetite and energy expenditure.

Gut regulation

Various factors influence food intake: nutritional status, smell, taste and a number of gut-produced peptide hormones (Table 62.1). These peptides communicate with the hypothalamus to control short-term appetite and satiety.

Cholecystokinin

Cholecystokinin (CCK) is mainly released from the duodenum and jejunum in response to eating, and causes gallbladder contraction and decreased gastric emptying. In the hypothalamus, CCK reduces food intake.

Pancreatic polypeptide

Pancreatic polypeptide is mainly produced in the endocrine pancreas and is released in response to eating. It too has
appetite-reducing effects. Peptide YY belongs to the same group of proteins as pancreatic polypeptide and is synthesised in the L-cells of the gastrointestinal tract. It is released in response to food intake but levels are lowered in obesity.

Glucagon-like peptide-1

GLP-1 is also released by intestinal L-cells in response to eating, acting to reduce appetite, gastric emptying and gastric acid secretion, and stimulate insulin secretion. GLP-1 analogues have found a therapeutic role in the treatment of T2DM, and their weight-reducing properties make them potentially useful as therapies for obesity.

Glucose-dependent insulinotrophic peptide

GIP is produced by the K-cells of the duodenum and jejenum, and is released in response to a high intestinal glucose load. It has similar incretin-like effects to GLP-1, inducing pancreatic insulin secretion in response to a meal.

Oxyntomodulin

Oxyntomodulin is a similar peptide to and released alongside GLP-1 in response to food. Oxyntomodulin therefore reduces food intake, increases energy expenditure and weight loss.

Ghrelin

In contrast to the above, ghrelin, a peptide hormone produced mainly in the stomach, acts to stimulate appetite. This peptide binds to hypothalamic receptors to exert its effects on the NPY and AgRP appetite-stimulating neurons.

Long-term regulation

Leptin

Leptin is a hormone produced in white adipose tissue, with the circulating concentration being proportional to the amount of body fat. It acts centrally in the hypothalamus (stimulating POMC neurons and inhibiting NPY/AgRP neurons) to reduce food intake and body weight, and to increase energy expenditure. However, obesity is associated with leptin resistance.

Insulin

Insulin can also act as a long-term regulator of food intake. It has similar effects on the hypothalamus, inhibiting the expression of appetite-stimulating NPY/AgRP neurons, and reducing appetite, food intake and weight. It also stimulates the production of leptin.

Miscellaneous factors

A number of other factors are thought to have a role in the long-term control of appetite and weight, including genetic factors, physical activity, psychological factors, food availability, cost and type.

Clinical assessment

The assessment of a patient’s appetite and weight requires a thorough history and examination.

History

The history should cover body systems whose disruption might lead to a change in appetite or weight, such as gastrointestinal (e.g. inflammatory bowel disease, malabsorption, infections), endocrine (e.g. hypo- or hyperthyroidism) or mood and behavioural symptoms (e.g. depressive disorders). Any inherited disorder affecting hypothalamic function or hypothalamic injury should be elicited. A past history of gastrointestinal surgery can affect gut hormone production, gastric emptying and appetite. Other chronic illnesses affecting appetite (e.g. malignancy, chronic kidney disease, dementia, HIV) should be noted as well as acute systemic illness (e.g. sepsis) that could have an impact on food intake and weight. In addition, any drugs with the potential to affect appetite and weight, including antibiotics, glucocorticoids, opiates and chemotherapy, should be noted.

Examination

The patient should be examined for any features of endocrinopathy, such as features of hypo- or hyperthyroidism,
adrenal insufficiency or Cushing’s syndrome. Any scars from previous surgery should be noted. The weight and height of the patient should be measured, with BMI calculated. BMI is defined as the weight in kilograms divided by the square of the height in metres (kg/m2) (Figure 62.1a). The BMI WHO categories are as follow (Figure 62.1b):

• <18.5 kg/m² – underweight
• 18.5–24.9 kg/m² – normal
• ≥25.0 kg/m² – overweight
• ≥30.0 kg/m² – obese.
  Waist : hip ratio (WHR) can also be calculated as waist circumference divided by hip circumference. WHR may more
  accurately reflect abdominal (visceral) fat than BMI. Ratios0.85 in women and >0.90 in men indicate abdominal obesity. Higher WHR, or even waist circumference alone, is associated with increased cardio-metabolic risk.

Obesity and anorexia

Obesity

Definition

Obesity is defined as a BMI >30 kg/m².

Epidemiology

accounting for around 15% of the world’s population. In the UK, approximately 25% of the adult population are obese, with a further 35% classed as overweight. Obesity rates are continuing to rise, such that obesity is estimated to affect 60% of adult men, 50% of adult women and 25% of children in the UK by 2050.

Aetiology

Simplistically, obesity can be viewed as an energy imbalance between energy consumed and energy expended. However, its aetiology is complex and a result of both genetic and environmental factors.

Genetic factors

Rare genetic disorders associated with obesity include Prader– Willi, Laurence–Moon–Biedl and Bardet–Biedl syndromes. However, single gene defects account for only a very small proportion of patients with obesity; for the most part, genetic predisposition is polygenic.

Environmental factors

Environmental factors include the increased availability, quantity and intake of high sugar and high fat content foods, coupled with a decrease in physical activity. Psychological and social factors include altered eating behaviour, lack of money to buy healthy foods and availability of places to exercise. In addition, certain medications, such as antidepressants, anticonvulsants, contraceptives, corticosteroids and insulin can all contribute to weight gain.

Clinical features

Patients with obesity often present with complications of obesity, including osteoarthritis, obstructive sleep apnoea, gallstones, T2DM, hypertension and cardiovascular disease (Figure 63.1a). Obesity can be diagnosed by calculating the BMI or measuring waist circumference. Endocrine disease is rarely causative, but an examination to look for signs of Cushing’s syndrome or hypothyroidism should be undertaken, as well as a search for features of the rare monogenic causes.

Investigations

A suspicion of endocrinopathy should lead to appropriate endocrine testing. Patients should be screened for T2DM
(Chapter 43) and a fasting lipid profile performed. Any suspected complications related to obesity should be investigated in the standard manner (e.g. ultrasound abdomen in gallstones).

Management

Managing obesity presents a huge challenge for the global healthcare community. Prevention is crucial but public health campaigns to date have failed to impact significantly on this growing epidemic.
Patients with established obesity should target at least a 10% weight loss as this is associated with significant reduction in morbidity and mortality. Dietary strategies aimed at reducing energy intake should be used, in addition to increasing the amount of physical activity.

Drug therapies

Drug therapies available to treat obesity are limited. Orlistat, an inhibitor of pancreatic and gastric lipases, can result in a modest reduction in weight of up to 10%. However, treatment is often poorly tolerated as a result of steatorrhoea from fat malabsorption. GLP-1 receptor agonists may have a future role as they are currently known to induce significant weight loss in many patients with T2DM.

Surgery

to reduce weight significantly in the long-term. Restrictive (gastric banding or sleeve gastrectomy) or malabsorptive (gastric bypass) procedures can be undertaken, but surgery in the UK is currently restricted to patients with a BMI of 40 kg/m2 or more, or 35–40 kg/m2 if significant co-morbidity (e.g. T2DM or hypertension) potentially amenable to improvement with weight loss is present (Figure 63.1b). All other non-surgical measures must have been tried first. Bariatric surgery is associated with a resolution of newly established T2DM in up to 80% of cases, hence patients with a recent diagnosis of T2DM and BMI ≥35 kg/m² can be assessed for surgery.

Prognosis

It is estimated that 25% of the ischaemic heart disease burden, 45% of the diabetes burden and up to 40% of certain cancers are caused by overweight and obesity. At least 2.8 million adults die each year as a result of being overweight or obese.

Anorexia

Definition

Anorexia is an eating disorder associated with a BMI <17.5 kg/m².

Epidemiology

Anorexia affects around 2 million people worldwide with a female to male ratio of 10 : 1.

Aetiology

Genetic factors are thought to play a part in the development of anorexia nervosa. There are also a number of psychological, social and emotional factors that contribute, including a family history of depressive disorder, low self-esteem, higher social class or stressors during adolescence.

Symptoms and signs

Symptoms of anorexia include deliberate weight loss, a fear of fatness and altered perception of body weight, accompanied by behavioural changes and amenorrhoea. Patients also display fatigue, irritability and coldness. Signs include a BMI <17.5 kg/m², scanty pubic hair, poorly developed breasts, lanugo hair, bradycardia and peripheral oedema (Figure 63.2a).

Investigations

Endocrine disruption can be widespread. Gonadotrophins (LH, FSH) and oestradiol are typically low. Prolactin is usually normal but cortisol may be elevated. Thyroid function can show a normal TSH with low-normal T4 (‘sick euthyroidism’). Hypomagnesaemia, hyponatraemia, hypoglycaemia, hypokalaemia and hypocalcaemia may be present, in keeping with the poor nutritional state (Figure 63.2b). A DXA scan can reveal osteopenia or osteoporosis (in part caused by low oestrogen).

Management

Treatment centres on early weight gain, use of antidepressants or antipsychotics, psychotherapy, cognitive behavioural therapy and family support (Figure 63.2c). Weight gain is the most important measure to restore normal gonadal function but in cases of osteoporosis, oestrogen therapy can be required.

Prognosis

Around 60% of patients relapse, 20% make a good recovery and 20% have a poor outcome with high mortality.