Type 2 diabetes mellitus
Type 2 diabetes has achieved pandemic proportions and affects some 300 million people worldwide. It arises because insulin production is inadequate relative to the demands of the body, which may be increased by obesity or other factors. Its dramatic increase and earlier onset over the past century reflect increased calorie intake, less exercise and other lifestyle factors. Type 2 diabetes is heterogenous and reflects the interaction of several disease processes. It therefore lacks a precise definition, and its clinical features vary widely within and between populations. It is best considered as a syndrome rather than a disease. Measures such as diet and weight loss that reduce the demand for insulin may delay or reverse the metabolic abnormality, but beta cell failure is typically progressive, with declining insulin production and increasing reliance on medication. Lifestyle change is the mainstay of treatment, backed by a range of oral therapies, although patients frequently progress to insulin. Glucose control has been shown to reduce the impact of small vessel complications of diabetes together with cataracts and neuropathy, but has less effect upon arterial disease, the leading cause of premature death in type 2 diabetes. Careful attention to other cardiovascular risk factors is therefore needed.
Many 19th century clinicians knew that there were two forms of diabetes. Young, thin patients became acutely ill and soon died from uncontrolled diabetes. Older overweight individuals, in contrast, often lived for years with the help of a strict diet. The discovery of insulin showed that the first type of patient relied on insulin for health and survival, whereas insulin was not essential for the remainder; the two varieties were sometimes called juvenile-onset and maturity-onset diabetes. Tablet treatments such as tolbutamide became available in the 1950s, and the terms insulin-dependent and non-insulin-dependent diabetes (IDDM and NIDDM) came into use. The immune pathogenesis of early onset diabetes was recognised in the 1970s, and it was acknowledged that type 1 and type 2 diabetes, as they are now called, were effectively two different diseases. The current terminology was formally adopted in the 1990s. The causes of type 2 diabetes are unknown, and the term is often used to refer to any form of diabetes of unknown aetiology that does not meet the criteria for type 1 diabetes. Further subtypes will undoubtedly come to be recognised, but it in the interim it is important to recognise that type 2 diabetes takes many forms in different environments and against differing genetic backgrounds.
Diabetes was considered uncommon in the 19th century, partly because diagnosis relied on tasting the urine, and its relationship to obesity was not fully appreciated until the 1920s. Diabetes was at that time considered to favour people of European extraction, but this was mainly because they enjoyed a more affluent lifestyle. Globalisation and economic development have created a worldwide epidemic of obesity and of type 2 diabetes, which is now more common among people of Asian, African or Hispanic American origin than among Europeans. Its prevalence is particularly striking in societies that have made a rapid transition from a traditional to an affluent lifestyle, a process sometimes referred to as 'Coca-Colonisation'. The rapid development of diabetes in previously undernourished populations gave birth to the 'thrifty gene' hypothesis, which postulated that genes predisposing to diabetes are helpful when food is in short supply. Attractive though the hypothesis undoubtedly is, this has never been confirmed.
Major risk factors for type 2 diabetes other than obesity are family history, age and ethnicity. Women gained weight more rapidly than men in western societies over the first half of the 20th century, and were more subject to diabetes, but weight gain and overall prevalence of diabetes are now equal.
Genetics of type 2 diabetes
Type 2 diabetes often runs in families, although it is not always easy to distinguish between the effect of shared genes and shared environment. If one member of a twin pair develops diabetes, however, the risk of diabetes in the unaffected twin is much greater in monozygotic (identical) than in dizygotic (non-identical) twin pairs, underlining the importance of the genetic contribution. The extent to which the genes conferring susceptibility to type 2 diabetes will be expressed depends upon environmental factors such as obesity. Multiple genes are involved in genetic susceptibility, each making a small contribution, and their prevalence varies between populations. Many of these genes influence aspects of healthy beta cell function.The genes most strongly associated with type 2 diabetes are TCF7L2 in Western populations and KCNQ1 in Asian populations. TCF7L2 is associated with reduced insulin secretion and increased hepatic glucose production. Other important genes include SLC30A8, a zinc transporter involved with crystallisation of insulin and peroxisome proliferator-activated receptor (PPAR)γ, a receptor that controls the expression of several genes. The FTO gene increases diabetes susceptibility indirectly by contributing to weight gain. The current pay-off of genetic research in type 2 diabetes is limited by modest effect sizes (only 5–10% of susceptibility is currently explained), limited knowledge of the underlying biological pathways, and limited predictive ability. Improved clinical translation and personalized disease profiling and therapy remain important targets of research [McCarthy].
Pathogenesis of type 2 diabetes
Schematic to show that diabetes presents earlier in insulin resistant individuals, and that these have greater beta cell mass at diagnosisDiabetes, however caused, results when the beta cells of the pancreas are no longer able to meet the body’s requirement for insulin. Hyperglycaemia therefore develops earlier in the course of beta cell failure, and in association with a greater beta cell mass, in insulin-resistant compared with insulin-sensitive individuals (see Figure). The relative importance of reduced secretion of insulin and increased demand (insulin resistance) ranges along a spectrum from one individual or population to the next. Historically, insulin resistance was thought to cause type 2 diabetes, partly because insulin secretion is maintained or enhanced at diagnosis, and the beta cell mass appeared relatively intact at autopsy. It was later appreciated that beta cell mass is reduced in type 2 diabetes relative to body mass, and that insulin resistant individuals, e.g. the massively overweight, will remain non-diabetic if in possession of a healthy pancreas. Furthermore, as noted above, many of the genes associated with type 2 diabetes influence beta cell function. The causes and mechanism of beta cell dysfunction have thus become a major research focus.
The islets in type 2 diabetes
The pancreatic islets appeared normal to most early microscopists, although abnormal staining adjacent to the beta cells, variously described as hyaline (glass-like) or amyloid (starch-like) in appearance, was noted 100 years ago. These are now known to be due to accumulation of islet-associated amyloid peptide (IAPP), an insoluble peptide co-secreted with insulin. The role of IAPP, whether cause or consequence of diabetes, remains controversial. Experimental studies of islet function have largely been performed in rodent models, which respond to obesity by making more islets [Prentki], whereas adult human islets have limited regenerative capacity. Human beta cell mass and/or function decrease with age, as reflected in the fact that some 50% of octagenarians have dysglycaemia or frank diabetes. Possible explanations for the development of type 2 diabetes therefore include early formation of a beta cell mass inadequate for the demands of adult life, accelerated loss of beta cells with increasing age, factors limiting compensatory hypersecretion of insulin, or harmful consequences of such prolonged stress upon residual beta cells. Postulated mechanisms for initiation of beta cell failure include mitochondrial dysfunction, oxidative stress, endoplasmic reticulum (ER) stress, dysfunctional triacylglycerol/non-esterified fatty acid (TG/NEFA) cycling, and glucolipotoxicity. Secondary mechanisms which may promote islet damage at a more advanced stage include islet inflammation, O-linked glycosylation and deposition of IAPP. Better understanding of beta cell function in type 2 diabetes could delay or prevent initiation of progressive cell damage, might allow restoration of the intrinsic capacity of beta cells to replicate, or support better means of islet transplantation.
As noted earlier, the onset of diabetes is determined by the balance between insulin sensitivity and insulin secretion, and this is highly variable between populations. For example, Asian people with diabetes tend to be slimmer and more sensitive to insulin than Western people, and to have more advanced beta cell deficiency. As in type 1 diabetes, early signs of a failing beta cell mass include loss of pulsatile insulin secretion and loss of the first phase insulin response to intravenous glucose (FPIR). This is followed by a declining ability to dispose of a glucose challenge and increasing fasting plasma glucose or HbA1c. Insulin secretion is often increased relative to non-diabetic individuals at the onset of diabetes, although insufficient to restore normoglycaemia. Beta cell function then declines with increasing duration of diabetes, requiring progressive escalation of therapy, and often culminating in insulin treatment. There is however considerable variation in the rate at which beta cell function is lost. Oversecretion of glucagon is also present and exacerbates the consequences of insulin deficiency; it has been argued that type 2 diabetes is a bi-hormonal disorder.
Special tests of insulin secretion and sensitivity are available, but are rarely needed in routine clinical practice. Circulating insulin can be measured in people who have not received insulin therapy (insulin antibodies would otherwise interfere with the assay), but direct measurement of C-peptide, a short peptide with a longer half-life, is generally preferred. C-peptide becomes detached when proinsulin is converted to insulin and is cosecreted in equimolar proportions with insulin, and therefore provides a useful marker of insulin secretion. Insulin resistance is more difficult to measure directly, but can for research purposes be measured by the hyperinsulinaemic–euglycaemic clamp technique, which measures the amount of glucose that has to be infused to match an insulin infusion in human volunteers. Simpler estimates of insulin sensitivity can, however, be made from simultaneous measurement of fasting glucose and insulin, which form the basis of two widely used tests, homeostatic model assessment (HOMA) and the quantitative insulin sensitivity check test (QUICKI). These have the advantage that they can be used in large populations.
Differential diagnosis of type 2 diabetes
Type 2 diabetes is essentially a diagnosis of exclusion, and it is therefore important to consider the possibility of other causes of diabetes in newly presenting patients, since this may have an impact on the line of therapy. Other causes include endocrine disease, damage to the pancreas, iron storage disease, drug therapy and genetic syndromes [link]. Type 2 diabetes appears to travel as an autosomal dominant trait in some families, and maturity onset diabetes of the young (MODY) should be considered when diabetes presents early. Type 1 (autoimmune) diabetes may present at any age, and may develop more slowly in older people. A proportion of people presenting with apparently typical type 2 diabetes have markers of autoimmunity (glutamic decarboxylase antibodies, GADA, is the most frequently used marker) and these progress more rapidly to insulin following diagnosis. This is referred to as latent autoimmne diabetes in adults (LADA). One problem with this diagnosis is the high rate of false positives (about 50% of older patients with GAD antibodies would be expected to have them even if they did not have diabetes), and there is controversy as to its clinical implications. Some clinicians believe that the presence of GAD antibodies is an indication for early insulin treatment, while others maintain that this will make no difference to the outcome.
Therapy of type 2 diabetes
Lifestyle is strongly associated with the development of diabetes, and lifestyle change is the cornerstone of its management. Most people with an affluent lifestyle eat slightly more than they need, and their risk of diabetes can be reduced by eating less and/or exercising more. Control of energy intake and body weight is therefore the first principle of therapy, whereas there is still some controversy as to the importance of qualitative changes in diet, e.g. the proportion of energy derived from carbohydrate. Regular exercise is another basic element of treatment, but should be appropriate to the individual’s circumstances.
Many classes of drug are available for the treatment of diabetes. There are intestinal enzyme inhibitors that limit the uptake of carbohydrate (acarbose) or fat (orlistat) from the gut, insulin sensitisers that reduce the body’s requirement for insulin (metformin, thiazolidinediones), insulin secretagogues (sulfonylureas, meglitinides) that increase its ability to produce insulin, and a new class of agents, the SGLT2 inhibitors, that promote loss of circulating glucose into the urine. Other potentially useful drug effects include inhibition of glucagon secretion and delayed gastric emptying, and many drugs for diabetes have multiple actions. As yet, and despite claims to the contrary, no agents have convincingly been shown to prolong beta cell life to any useful extent or to induce beta cell regeneration in humans.
An alternative (or complementary) approach to therapy uses the gut hormone glucagon-like peptide-1 (GLP-1). GLP-1 is secreted by the L-cells of the gut wall, but rapidly inactivated by the enzyme dipeptidyl peptidase (DDP)4. GLP-1 based therapies act by inhibition of DPP4, thereby increasing availability of GLP-1 to the body, or can be given by injection as GLP-1 analogues, molecules that resemble native GLP-1 in their action but are less readily broken down in the circulation. The longer term safety of these agents is still under consideration.
Since the ultimate aim of all therapies for diabetes is to prolong the life and well-being of their users, overall safety, acceptability and cost become major factors when choosing a therapy. Rosiglitazone, for example, had useful glucose-lowering effects but appeared to increase cardiovascular risk and was therefore withdrawn from general use.
Prevention of type 2 diabetes
Prevention may be primary (before onset of disease-related abnormalities), secondary (when early signs of disease are present) or tertiary (after diagnosis).
During World War II, in the 'Hunger winter' of 1944 food shortage in the Netherlands led to a dramatic drop in the incidence of diabetesBoth the incidence and mortality of diabetes fell during the Second World War in populations subject to food rationing, and a healthy diet, controlled calorie intake and regular exercise form the basis of primary prevention. Secondary prevention has been tested in individuals with abnormal glucose tolerance. Lifestyle measures proved most effective, followed by use of metformin or acarbose. Other oral agents have been shown to lower glucose levels in this situation, but not to affect the course of the disease. Obesity-related diabetes can be reversed temporarily by weight loss, and this is particularly marked following bariatric surgery.