Caballero B. (ed.) Encyclopaedia of Food Science, Food Technology and Nutrition. Ten-Volume Set

Подождите немного. Документ загружается.

DIABETES MELLITUS

Contents

Etiology

Chemical Pathology

Treatment and Management

Problems in Treatment

Secondary Complications

Etiology

J Wright, University of Surrey, Guildford, Surrey, UK

Definition and Classification

0001 Diabetes mellitus is a complex disorder of metabol-

ism (fat and protein as well as carbohydrate) caused

by either a deficiency of, or defective action of,

insulin. It is not a single disease entity, but rather a

syndrome of variable clinical presentation and

etiology (Table 1). (The terms insulin-dependent dia-

betes mellitus (IDDM) and noninsulin-dependent

diabetes mellitus (NIDDM) should now be aban-

doned in favor of type 1 and type 2 diabetes.)

Basic Features of Diabetes

0002 Type 1 (insulin-dependent) and type 2 (noninsulin-

dependent) diabetes account for the great majority

of patients. They differ in terms of their clinical

presentation, etiology, and basic metabolic defect.

Type 1 diabetes is characterized by absolute insulin

deficiency owing to B-cell destruction, and type 2

diabetes by relative insulin deficiency due to a

combination of dysfunctional insulin secretion and

resistance to insulin.

0003Successful treatment of diabetes with diet, oral

hypoglycemic drugs, or insulin enables the majority

of patients to lead long and near-normal lives. In-

creasingly, it is the long-term complications of the

disease which are the major challenge in the treat-

ment of diabetes. The most important of these are the

microvascular complications (retinopathy and neph-

ropathy), the macrovascular complications (athero-

sclerosis, coronary artery, and peripheral vascular

disease), and neuropathy. These complications are

uncommon in type 1 diabetes of less than 10 years’

duration but are not uncommon at presentation in

type 2 diabetes, probably because of the long period of

asymptomatic hyperglycemia which usually precedes

the clinical diagnosis. In the majority of patients it is

relatively easy to achieve good or at least reasonable

health in the short term. However, with increasingly

convincing evidence that the incidence of long-term

complications can be reduced by maintaining good

glycemic and metabolic control, this has become the

major emphasis in the clinical monitoring and man-

agement of diabetes.

0004The major differences between types 1 and 2

diabetes are shown in Table 2.

tbl0001 Table 1 Etiological classification of diabetes

Type 1: insulin-dependent diabetes mellitus (IDDM)

Autoimmune

Idiopathic

Type 2: noninsulin-dependent diabetes mellitus (NIDDM)

Other specific types

Genetic defects of B-cell function

Genetic defects in insulin action

Diseases of the exocrine pancreas

Endocrinopathies

Drug- or chemical-induced

Infections

Uncommon forms of immune-mediated diabetes

Other genetic syndromes sometimes associated with diabetes

Gestational diabetes

The terms IDDM and NIDDM should now be abandoned in favor of type 1

and type 2 diabetes.

Based on American Diabetic Association Expert Committee Report (1997).

tbl0002Table 2 Major features of types 1 and 2 diabetes

Type1diabetes Type 2 diabetes

Age at onset (years) Usually <30 Usually > 40

Weight Weight loss Normal or

overweight

Rate of onset Rapid Slow

Ketosis Usual Absent

Complications at presentation Rare Common

HLA association Usual Absent

Islet cell antibodies Usual Absent

HLA, human leukocyte antigen.

Adapted from Home PD, Johnston DG and Alberti KGMM (1989) Diabetes

mellitus. In: Hall R and Besser G (eds) Fundamentals of Clinical

Endocrinology, pp. 318–361. Edinburgh: Churchill Livingstone.

1776 DIABETES MELLITUS/Etiology

0005 Clinical diabetes represents a continuous spectrum

from the young, underweight, ketotic, and sometimes

comatose patient on the one hand to the elderly, obese

and frequently asymptomatic patient on the other.

The majority of patients with types 1 and 2 diabetes

conform to the patterns in Table 2 but there are

frequent exceptions; it is increasingly recognized, for

example, that the elderly sometimes develop typical

type 1 diabetes. There is more uniformity in young

patients, almost all of whom require insulin. An

exception is the syndrome of maturity-onset diabetes

of the young (MODY), a group of rare inherited

disorders which usually present before the age of

25 years, and which, in the first instance at least,

do not usually require treatment with insulin.

Diagnosis

0006 Although both insulin deficiency and insulin resist-

ance result in abnormalities of fat, protein, and carbo-

hydrate metabolism, it is upon the changes in blood

glucose levels that both diagnosis and monitoring are

primarily based.

0007 Until recently, the diagnosis of diabetes was based

upon the criteria set by the World Health Organiza-

tion (WHO), using a standard oral glucose tolerance

test (OGTT). In the vast majority of patients, a GTT

is unnecessary, and the diagnosis is made on the basis

of clinical history, and the random and/or fasting

blood glucose level, which are usually unequivocal.

(The exception is pregnancy, during which screening

for gestational diabetes using a GTT is common prac-

tice.) However, strict criteria and a standard diagnostic

procedure are important, particularly for epidemi-

ological purposes. The interpretation of a GTT is

based on the glucose level in the fasting state and 2 h

after a 75-g glucose load (or equivalent); neither the

1-h glucose level nor measurements of urine glucose

or plasma insulin are of diagnostic use. The glucose

load should be dissolved in 250–300 ml of water and

drunk within 5 min after an overnight fast; dietary

carbohydrate should not have been restricted during

the 3 days prior to the test.

0008The WHO criteria have recently been revised, and

the new criteria are shown in Table 3. The major

reason for the changes is the fact that the 1-h postload

plasma glucose level of 11.1 mmol l

1

(the most im-

portant criterion for the diagnosis of diabetes) correl-

ates better with a fasting level of 7 than 6 mmol l

1

new category of ‘impaired fasting glucose’ (fasting

plasma glucose 6.1–7.0 mmol l

1

) has also been pro-

posed because it has been shown that prevalence of

diabetic complications (retinopathy) starts to increase

significantly in people with a fasting plasma glucose

above 6 mmol l

1

.(See Glucose: Maintenance of

Blood Glucose Level; Glucose Tolerance and the

Glycemic (Glycaemic) Index.)

0009Impaired glucose tolerance (IGT) is not a separate

etiologic category. It is associated with an increased

risk of developing diabetes (10–25% within 5 years)

and, importantly, the risk of developing the syndrome

of obesity, hypertension, and hyperlipidemia as well

as the macrovascular complications of diabetes.

Individuals with IGT should therefore be monitored

carefully, and treated with dietary restriction and

weight loss where appropriate. (See Hyperlipidemia

(Hyperlipidaemia); Hypertension: Nutrition in the

Diabetic Hypertensive.)

Epidemiology

0010The current and predicted prevalence of diabetes

worldwide is shown in Table 4. As can be seen, it is

anticipated that the prevalence of diabetes will (ap-

proximately) double by the year 2010, with the major

increase occurring in ‘developing’ countries. Around

98% of the increase is accounted for by type 2 diabetes.

Type 1 Diabetes

0011There are dramatic differences in the prevalence of

type 1 diabetes, both racially and geographically. The

tbl0003 Table 3 Diagnostic values for the oral glucose tolerance test

(WHO, 1999)

Venous plasma glucose

(mmol l

1

)

Diabetes mellitus

Random, or 11.1

Fasting, or 7.0 (whole blood 6.1)

2 h after glucose load 11.1

Impaired glucose tolerance (IGT)

Fasting, and < 7.0

2 h after glucose load 7.8 but 11.1

Impaired fasting glycemia (IFG)

Fasting 6.1 but < 7.0

tbl0004Table 4 Estimated prevalence of diabetes (millions)

Worldwide 1997 124 m

2010 221 m

West Asia 1995 3.6 m

2010 11.4 m

South/central Asia 1995 28.8 m

2010 57.5 m

South-east Asia 1995 8.6 m

2010 19.5 m

East Asia 1995 21.7 m

2010 44 m

2% population; 3.5 m type 1, 120 m type 2.

From Amos AF, McCarty DJ and Zimmet P (1997) The rising global burden

of diabetes: estimates and projections to the year 2010. Diabetic Medicine

14: S7–S15.

DIABETES MELLITUS/Etiology 1777

annual incidence rate in young people below the age

of 20 years varies from around 30 per 100 000 per

year in Finland and Scandinavia to less than 1 per

100 000 per year in oriental populations such as the

Koreans and the Japanese. Rates are generally much

higher in northern than southern Europe (with the

exception of the island of Sardinia, which has an

incidence similar to Scandinavia), and higher in

whites than blacks from the same geographical area.

Racial differences in the prevalence of type 1 diabetes

may relate to differences in the frequency of suscepti-

bility genes in the population.

0012 The peak age of onset of type 1 diabetes is between

the ages of 11 and 13 years. Thereafter, the incidence

rate is low during middle life but there is a further

peak in the elderly, in whom the presenting features of

the disease may be similar to those seen in children.

The etiology in this group may be different, but they

frequently have islet-cell antibodies, indicating an

autoimmune process. In addition to age-related dif-

ferences in incidence, there is also a seasonal vari-

ation, seen only in older children, with a maximum

during the winter months, possibly related to seasonal

variations in viral illnesses involved in the etiology of

the disease. Of particular concern is the fact that the

incidence in children appears to be increasing steadily

in some countries, including the UK. The reasons for

this are unclear.

Type 2 Diabetes

0013 There are also marked differences in the prevalence

and incidence of type 2 diabetes. The lowest rates are

found in underdeveloped rural communities in Africa

and Asia. The prevalence in most European countries

is between 1% and 3%, and it is estimated that there

may be a further similar number of cases which are

undiagnosed. Unlike type 1 diabetes, surveys in the

USA show that the prevalence of type 2 diabetes is

higher among blacks than among whites.

0014 The anticipated explosion in the prevalence of

type 2 diabetes is already evident in certain ethnic

groups who have been subjected to rapid ‘westerniza-

tion.’ The most extensively studied are the Pima

Indians in Arizona and the Micronesian population

of the Pacific island of Nauru where inheritance

appears to be autosomal dominant, and the preva-

lence in adults may be as high as 80%. Recently, high

rates have also been found in a number of migrant

populations, particularly in those from the Indian

subcontinent to the west. It has been suggested that

these peoples possess a ‘thrifty genotype’ which con-

fers the advantage of metabolic efficiency when food

is in short supply but which leads to obesity, insulin

resistance, and type 2 diabetes when food becomes

plentiful.

Complications and Mortality

0015The pattern of complications varies between type 1 dia-

betes and type 2 diabetes. In both, the major cause of

death is from cardiovascular disease, but in type 1 dia-

betes, microvascular complications (nephropathy and

retinopathy) remain important, with a significant in-

crease in deaths from renal disease and its sequelae.

Similar microvascular complications occur in type 2

diabetes, but the increased mortality is largely attribut-

able to deaths from cardiovascular disease. These dif-

ferences probably reflect fundamental differences in the

pathophysiology of the two conditions.

Causes

0016The underlying lesions in type 1 and 2 diabetes are

different and result from different pathogenic pro-

cesses. In the majority of cases of type 1 diabetes

there is loss of insulin secretion due to autoimmune

destruction of B cells, while in type 2 diabetes there is

usually minimal loss of B cells but evidence of dys-

functional insulin secretion and peripheral insulin

resistance. In both cases the disease process is com-

plex and almost certainly involves genetic, environ-

mental, and possibly dietary factors.

Type 1 Diabetes

0017Genetic factors About 60% of the genetic predis-

position to autoimmune type 1 diabetes is accounted

for by genes of the major histocompatibility complex

(MHC), particularly the human leukocyte antigen

(HLA) genes which code for a series of cell-surface

glycoproteins, class I and class II HLA antigens. Class

I proteins are expressed on the surface of most nucle-

ated cells and are involved in the presentation of

processed antigens to cytotoxic T lymphocytes.

Class II proteins are expressed on B lymphocytes,

macrophages, and activated T cells, and are involved

in the presentation of antigens to T-helper lympho-

cytes. Class II proteins comprise a and b chains which

are encoded for by genes within the DP, DQ, and DR

regions of chromosome 6.

0018The major association with type 1 diabetes is with

possession of the HLAs DR3 and DR4 (Table 5).

tbl0005Table 5 Human leukocyte antigen (HLA) associations and the

relative risk of type 1 diabetes

HLA-DRantigen Relative riskof

type 1diabetes

95% confidence limits

DR2 0.1 0.05–0.3

DR3 5.0 2.9–8.8

DR4 6.8 3.8–12.1

DR3 and DR4 14.3 6.3–32.4

Data from the Barts-Windsor Study (1983).

1778 DIABETES MELLITUS/Etiology

However, these are too frequent in the general popu-

lation to be solely responsible for susceptibility to

diabetes, and the reason that these haplotypes appear

to confer susceptibility is that they are in linkage

disequilibrium (i.e., transmitted together) with other

true susceptibility alleles, particularly within the DQ

region. One particular DQ polymorphism determines

the presence of an aspartate residue at position 57 in

the DQ8 b chain. The presence of an aspartate residue

in this position changes the configuration of an

antigen-presenting cleft in the DQ protein; aspartate

confers a more efficient antigen-presenting configur-

ation, and thus enhances antigen recognition and the

subsequent immune response. In addition to confer-

ring susceptibility, the possession of certain HLA

alleles may confer protection (e.g., DQ6, DQ18).

0019 While a considerable amount is known about the

MHC/HLA genes, very little is known about the

genes which may account for the remaining 40% of

genetic susceptibility. There is some evidence that the

insulin gene may be a susceptibility locus, but conclu-

sive evidence is lacking.

0020 Environmental factors Twin studies have shown

that the concordance for type 1 diabetes between

genetically identical individuals may be less than

40%, indicating that other environmental factors

are also involved in triggering the autoimmune pro-

cess. The most likely of these is viral infection, and

several candidate viruses have been identified. The

best documented is congenital rubella, with type

1 diabetes developing in 10–20% of affected individ-

uals. Others include mumps, and coxsackie B, cyto-

megalovirus, and Epstein–Barr virus. There may be

several mechanisms by which viral infections may

cause B-cell damage. These include direct B-cell infec-

tion and damage, induction of antigenic determinants

in the B-cell membrane, either as a result of incorpor-

ation of viral proteins or exposure of autoantigens,

and molecular mimicry (immunological cross-reactiv-

ity between viral antigens and B-cell components).

0021 Other possible environmental triggers include toxic

and/or nutritional factors. Nitrosamines in smoked

mutton have been implicated in the pathogenesis of

childhood diabetes in Iceland. Currently there is con-

siderable concern that bovine serum albumin (BSA) in

cows’ milk may be able to initiate an autoimmune

response because of cross-reactivity between BSA and

a B-cell component; this issue is clearly of major

importance but the connection remains unproven.

0022 Immune factors It is likely that type 1 diabetes

develops as the result of an environmental injury in

genetically predisposed individuals, the combination

of which initiates an autoimmune process which

results in insulitis and B-cell death. The precise details

of this process are unclear but it is probable that an

initial event such as a viral infection results in B-cell

damage and exposure of antigens which are usually

hidden from the immune system. This triggers a

macrophage response and subsequent release of cyto-

kines such as tumor necrosis factor and g-interferon

which are capable of inducing aberrant expression of

class 2 antigens on B cells, thus enabling B cells to act

as antigen-presenting cells, and to present surface

autoantigens directly to T cells. Neither A nor D

cells appear to be capable of being induced to express

class 2 antigens and are therefore spared.

0023The result of the autoimmune process is the pro-

duction of autoantibodies which can be detected in

the circulation of almost all newly diagnosed cases of

type 1 diabetes. Antibodies may be directed against

either cytoplasmic antigens (islet-cell antibodies,

ICAS) or against surface antigens (islet-cell surface

antibodies, ICSAS), and even against insulin itself

(insulin autoantibodies, IAAS). They may be comple-

ment-fixing and therefore cytotoxic. Candidate

cytoplasmic autoantigens include the enzyme glutam-

ate dehydrogenase (GAD), while candidate cell-

surface autoantigens include the glucose transporter

GLUT-2.

0024Autoantibodies may be present in the circulation

for months or even years before the clinical presenta-

tion of the disease, during which time insulin secre-

tion may decline gradually. There may then be a final

event which precipitates the development of clinical

diabetes, either a further viral infection or reexposure

to the original triggering infection, thus explaining

the seasonal peaks in the incidence of new cases of

type 1 diabetes.

0025Idiopathic type 1 diabetes In a small proportion of

patients, there are no genetic or autoimmune markers,

and the underlying cause (or possibly, causes) of dia-

betes in this group is unknown. This has been termed

idiopathic type 1 diabetes. It has been proposed that

type 1 diabetes should be subdivided into type 1A

(autoimmune) and type 1B (idiopathic) diabetes, but

this classification is not yet generally accepted.

Type 2 Diabetes

0026The major pathophysiological components of type 2

diabetes are dysfunctional insulin secretion and insu-

lin resistance. Indeed, in the majority of patients with

type 2 diabetes, the condition is part of a wider

syndrome – the ‘insulin resistance’ or ‘metabolic’ syn-

drome (previously known as syndrome X or Reaven’s

syndrome), the features of which are shown in

Table 6. The definition, nature, and assessment of

insulin resistance are described in the following

DIABETES MELLITUS/Etiology 1779

chapter (Diabetes Mellitus: Chemical Pathology). In-

sulin resistance arises from the interplay of a series of

genetic factors on the one hand, together with social

and environmental factors on the other.

0027 Genetic factors Genetic factors are clearly much

stronger in type 2 than in type 1 diabetes. In most

cases it is possible to obtain a family history, and

concordance is greater than 90% for type 2 diabetes

in monozygotic twins compared with around 40%

for type 1 diabetes. In high-risk populations, the risk

of diabetes in offspring when both parents have type 2

diabetes may be greater than 50%, and as high as

80% in Pima Indians and Nauru islanders when

parents develop the disease before the age of 45

years. Despite these figures, much less is known

about the genetic associations of type 2 than those

of type 1 diabetes. A number of genetic defects have

been identified which are responsible for rare diabetic

syndromes (Table 7) but in the great majority of

cases, inheritance is almost certainly polygenic, with

environmental factors playing a major role. There is

no evidence for association with MHC or HLA

markers, and only limited evidence for the involve-

ment of a number of other candidate genes including

the insulin gene itself, and insulin receptor and

glucose transporter genes.

0028 Social and environmental factors The major non-

genetic factors which predispose to the development

of insulin resistance are obesity, physical inactivity,

nutritional factors, and smoking.

0029 Increasing obesity is associated with deterioration

in glucose tolerance, and the majority of patients with

type 2 diabetes are overweight: around 75% have a

body mass index (BMI) of over 25 kg m

2

. Of course,

the fact that not all obese people develop diabetes is

evidence that other factors (particularly genetic pre-

disposition) are important. If obesity has a role in the

cause of type 2 diabetes, it is probably in modulating

the speed of progression to symptomatic disease, pos-

sibly by contributing to the development of insulin

resistance and the earlier onset of B-cell exhaustion.

The risk of developing type 2 diabetes is related to not

only the degree but also the distribution of obesity,

with the greatest risk seen with male-pattern central

obesity. (The precise determinants of fat distribution

are uncertain but it too is at least in part genetically

determined.) Central adipose tissue is more metabol-

ically active with higher lipolytic activity than periph-

eral fat, and it is thus a richer source of nonesterified

fatty acids (NEFA). Furthermore, the venous drainage

of visceral adipose tissue in particular goes directly to

the liver. High levels of NEFA stimulate hepatic

gluconeogenesis, impair peripheral (muscle) glucose

uptake, and inhibit hepatic degradation of insulin, all

of which predispose to the development of hypergly-

cemia and insulin resistance.

0030Although obesity is an important risk factor for

diabetes, both insulin resistance and the hyperinsuli-

nemia which results from it are predictive of diabetes

independently of body weight.

tbl0007Table 7 Other specific types of diabetes

Genetic defects of B-cell function

MODY1: HNF4a (chromosome 20)

MODY2: glucokinase (chromosome 7)

MODY3: HNF1a (chromosome 12)

MODY4: IPF-1 (chromosome 13)

Mitochondrial DNA 3243 mutation

Genetic defects in insulin action

Type A insulin resistance; leprachaunism

Lipoatrophic diabetes

Diseases of the exocrine pancreas

Pancreatitis, neoplasia

Trauma/pancreatectomy, etc.

Endocrinopathies

Cushing’s syndrome, acromegaly

Glucagonoma, pheochromocytoma

Drug- or chemical-induced

Nicotinic acid, glucocorticoids, thyroid hormone

a and b adrenergic agonists

Thiazide diuretics, phenytoin, ciclosporin

Ethanol

Infections

Congenital rubella, cytomegalovirus, etc.

Uncommon forms of immune-mediated diabetes

Insulin and insulin receptor antibodies

‘Stiff man’ syndrome

Other genetic syndromes sometimes associated with

diabetes

Down’s syndrome, Friedreich’s ataxia

Huntington’s chorea, Klinefelter’s syndrome,

Lawrence–Moon–Biedel syndrome, myotonic dystrophy, etc.

MODY, maturity-onset diabetes of the young.

Adapted from American Diabetes Association Expert Report (1997).

tbl0006 Table 6 Features of the insulin resistance (‘metabolic’)

syndrome

Impaired insulin-stimulated glucose uptake

Hyperinsulinemia

Central male-pattern obesity

Glucose intolerance or type 2 diabetes

Dyslipidemia*

. Increased plasma triglycerides

. Decreased HDL-cholesterol

. Increased small dense LDL particles

Hypertension

Hyperuricemia (gout)

Hypercoagulable state (increased PAI-1 and factor VII)

Polycystic ovary syndrome

Microalbuminuria

*The ‘Atherogenic Lipoprotein Phenotype’.

HDL, high-density lipoprotein; LDL, low-density lipoprotein; PAI-1;

plasminogen activator inhibitor-1.

1780 DIABETES MELLITUS/Etiology

0031 Physical exercise influences both glucose transport

systems (GLUT 4) and a number of enzyme systems

(e.g., glycogen synthase) in skeletal muscle which

have a marked effect on insulin sensitivity. The con-

tribution of smoking to the development of insulin

resistance is minor but significant.

0032 As far as diet is concerned, the relative roles of fat

and carbohydrate in the pathogenesis of insulin resist-

ance and diabetes is unclear and, indeed, controver-

sial. In many countries where the prevalence of

diabetes is low, dietary fat intake is also low and

dietary carbohydrate tends to be unrefined, with a

low glycemic index and a high proportion of soluble

and insoluble fiber, all of which mitigate against the

development of diabetes. However, overall energy

intake tends to be low in this situation and the preva-

lence of obesity is low. The situation in ‘developed’

countries is somewhat different. There is general

agreement that diets high in saturated fat predispose

to the development of insulin resistance, and as far as

overall macronutrient balance in the diet is con-

cerned, high-carbohydrate diets appear to increase

rather than decrease insulin resistance when com-

pared to diets high in mono- or polyunsaturated fat.

0033 Thrifty phenotype or thrifty genotype? The reason

that the incidence of insulin resistance and type 2

diabetes is so high in ‘developing’ societies has been

ascribed to the possession of a ‘thrifty’ gene which, in

an environment with an unstable, poor food supply,

confers a survival advantage due to metabolic effi-

ciency. When individuals with this genotype are ex-

posed to a copious western diet, metabolic efficiency

predisposes to weight gain, insulin resistance, and

diabetes.

0034 An alternative, or perhaps complementary hypoth-

esis (the thrifty phenotype) derives from an increasing

amount of evidence which indicates that nutrition in

utero and in early infancy influences insulin sensitiv-

ity in later life; babies who are small at birth and at

1 year of age tend to be more obese in later life, with

insulin resistance and a high risk of type 2 diabetes,

hypertension, and coronary heart disease. These two

hypotheses are clearly not mutually exclusive.

Other Specific Types of Diabetes

0035 In addition to the common patterns of type 1 and type

2 diabetes described above, IGT or diabetes may

occur in association with a large number of other

conditions, as shown in Table 7. The majority of

these conditions are very uncommon or, indeed,

rare. In the majority, the precise mechanism of the

carbohydrate intolerance is unclear but in some

(the various forms of maturity-onset diabetes of the

young (MODY), for example), identification of

the underlying biochemical defect has provided useful

insights into the mechanisms that may be involved in

the common polygenic form of type 2 diabetes.

Screening and Prevention

Type 1 Diabetes

0036Different arguments can be advanced for screening

for the different types of diabetes. Screening for sus-

ceptible individuals or for preclinical type 1 diabetes

might be justified as part of a prevention program

if means were available to avert or delay for a

significant time the onset of clinical disease. Once

glucose intolerance starts to develop, the progression

to clinical disease is usually rapid. Measurement of

blood or urine glucose is therefore useless, and

screening would, of necessity, be limited to testing

for susceptible genotypes or the presence of auto-

antibodies in children or young people with a family

history of the disease. Sporadic cases would, of

course, remain undetected, and since a high propor-

tion of patients do not have a family history, this

would clearly limit the effectiveness of screening.

Any such program would be extremely costly and

at present there is insufficient evidence that early

intervention might be effective. Measures such as

immunosuppressive therapy with ciclosporin or treat-

ment with nicotinamide have been shown either to

induce limited remission or to slow the rate of deteri-

oration of B-cell function. A further possible ap-

proach is to immunize high-risk individuals against

viruses capable of precipitating diabetes but this, too,

remains an unproven strategy.

0037Ultimately, preventive measures are likely to in-

volve gene therapy, but until both the technology

and ethical issues of such treatment have been

resolved, extensive screening for type 1 diabetes is

probably not justified.

Type 2 diabetes

0038Below the age of 45 years, the prevalence of type 2

diabetes in the population is over 3%, and screening

on a large scale is probably not cost-effective.

However, the nature of type 2 diabetes (frequently

asymptomatic), the high prevalence of diabetes (pre-

dominantly type 2 diabetes) in older age groups

(Table 8) and the high prevalence of complications,

both at presentation and in the early years after diag-

nosis, provide some justification for the establishment

of screening programs in the older population. Prob-

lems arise in deciding when to rescreen subjects who

were initially negative, since the pick-up rate will

clearly be lower on retesting. In addition to the

elderly, screening may be justified in those ethnic

DIABETES MELLITUS/Etiology 1781

groups in whom there is known to be a high incidence

of type 2 diabetes.

0039 To some extent, however, the question of establish-

ing specific screening programs is becoming redun-

dant as testing for diabetes forms part of a more

general health screen which general practitioners are

encouraged to provide.

0040 Screening can be based upon testing for glycosuria,

measuring fasting or random blood glucose levels or

upon measurements of glycated hemoglobin (HbA

Of these, blood glucose measurements are more

precise and probably to be preferred, provided that

they are interpreted in relation to the time of meals.

See also: Diabetes Mellitus: Chemical Pathology;

Glucose: Maintenance of Blood Glucose Level; Glucose

Tolerance and the Glycemic (Glycaemic) Index;

Hyperlipidemia (Hyperlipidaemia); Hypertension:

Nutrition in the Diabetic Hypertensive; Immunology of

Food

Further Reading

Amos AF, McCarty DJ and Zimmet P (1997) The rising

global burden of diabetes: estimates and projections to

the year 2010. Diabetic Medicine 14: S7–S15.

Daly ME, Vale C, Walker M, Alberti KGMM and Mathers

JC (1997) Dietary carbohydrate and insulin sensitivity: a

review of the evidence and clinical implications. Ameri-

can Journal of Clinical Nutrition 66: 1072–1085.

Expert Committee (1997) Report of the expert committee

on the diagnosis and classification of diabetes mellitus.

Diabetes Care 20: 1183–1197.

Harris MI, Hadden WC, Knowler WC and Bennett PH

(1987) Prevalence of diabetes and impaired glucose tol-

erance and plasma glucose levels in US population aged

20–74 yr. Diabetes 36: 523–534.

Home PD, Johnston DG and Alberti KGMM (1989) Dia-

betes mellitus. In: Hall R and Besser G (eds) Fundamen-

tals of Clinical Endocrinology, pp. 318–361. Edinburgh:

Churchill Livingstone.

Mahler RJ and Adler ML (1999) Type 2 diabetes mellitus:

update on diagnosis, pathophysiology and treatment.

Journal of Clinical Endocrinology and Metabolism 84:

1165–1171.

Pickup JC and Williams G (eds) (1997) Textbook of Dia-

betes, (2nd edn). Oxford: Blackwell.

Reaven GM (1988) Role of insulin resistance in human

disease. Diabetes 37: 1595–1607.

Singh BM, Rutter JD and Fitzgerald MG (1988) The natural

history of non-insulin-dependent diabetes mellitus.

Clinical Endocrinology and Metabolism 2: 343–358.

Wolf E, Spencer KM and Cudworth AG (1983) The genetic

susceptibility to type 1 (insulin-dependent) diabetes: an

analysis of the HLA-DR association. Diabetologia 24:

224–230.

Yki-Jarvinen (1997) MODY genes and mutations in

hepatocyte nuclear factors. Lancet 349: 516–517.

Chemical Pathology

J Wright, University of Surrey, Guildford, Surrey, UK

Morbid Anatomy of the Pancreas

The Normal Pancreas

0001Insulin is secreted from the B cells of the islets of

Langerhans. There are approximately one million

islets distributed throughout the pancreas. Individual

cells can be identified using immunocytochemistry,

whereby cells are stained using antibodies directed

at their secretory products. The B cell is the dominant

cell type of the islet but there are a number of others,

as shown in Table 1. There is a complex functional

and anatomical relationship between the cells of the

islet and their products. A and D cells are situated on

the periphery of the islet with a central core of B cells.

It is not entirely clear whether arterial capillary blood

reaching the B cells has first passed via both A and D

cells, transporting glucagon and somatostatin to the B

cells where they act as paracrine factors, respectively

enhancing and inhibiting the insulin response to stim-

uli, or whether capillary blood first supplies the B cell

tbl0008 Table 8 Age-related prevalence of diabetes in white

Caucasians

Age group (years) Prevalence of diabetesmellitus

0–9 0.5

10–19 1.8

20–29 4.0

30–39 4.6

40–49 7.5

50–59 13.8

60–69 21.3

70–79 37.1

80þ 35.6

Modified from Singh BM, Rutter JD and Fitzgerald MG (1988) The natural

history of non-insulin-dependent diabetes mellitus. Clinical Endocrinology

and Metabolism 2: 343–358.

tbl0001Table 1 Cell types and secretory products of the islet of

Langerhans

Cell type Secretory product

A Glucagon

B Insulin

D Somatostatin

PP Pancreatic polypeptide

1782 DIABETES MELLITUS/Chemical Pathology

with insulin, then exerts a paracrine effect on gluca-

gon and somatostatin secretion. The function of pan-

creatic polypeptide is not known.

0002 In addition to insulin, B cells also secrete amylin

(also known as islet amyloid polypeptide), a 37-

amino-acid peptide which is similar to calcitonin

gene-related peptide (CGRP, *50% sequence hom-

ology). It is found in secretory granules and appears

to be cosecreted with insulin. Initial studies suggested

that amylin may induce insulin resistance in skeletal

muscle. Whether altered amylin secretion contributes

to the metabolic abnormalities of diabetes is as yet

unclear. In type 1 diabetes, amylin production is lost

as a result of B-cell damage. In type 2 diabetes, amylin

secretion is diminished and the peptide accumulates

as amyloid fibrils which are found between the B cells

and the adjacent capillaries; there is some evidence

that this may impair insulin secretion and play a role

in the pathogenesis of type 2 diabetes.

Type 1 Diabetes (IDDM)

0003 In the majority of cases, type 1 diabetes is caused by

autoimmune destruction of B cells, probably triggered

by a viral infection in genetically susceptible individuals

(See Diabetes Mellitus: Etiology). The first visible

stages in this process involve the appearance of macro-

phages within the islets and the establishment of a

typical chronic inflammatory reaction with lympho-

cytic infiltration; this is termed ‘insulitis.’ Loss of B

cells is a gradual process which may first be detectable

several years before the development of clinical dia-

betes. In some patients, residual B-cell function persists

for many years after the onset of diabetes, with persist-

ence of insulitis in remaining B cells. Continuing insulin

secretion (as assessed by a positive C-peptide response

to glucagon stimulation) is characteristically associated

with easier glycemic control.

0004 In established type 1 diabetes, there is almost total

loss of B cells. As a result, the islets appear atrophic,

although the number and distribution of A, D, and PP

cells remain almost normal.

Type 2 Diabetes (NIDDM)

0005 Unlike type 1 diabetes, the gross morphology of the

islet is normal (with the exception of islet amyloid

accumulation). The extent of B-cell loss is variable

and usually only minimal, although there is occasion-

ally a loss of up to 50% of B cells; the number of A

and D cells remains unaffected.

The Metabolic Effects of Insulin

Insufficiency

0006 The major stimulus to insulin secretion is an increase

in blood glucose concentration. A number of other

physiological factors also stimulate insulin secretion

(Table 2), in particular the hormones of the entero-

insular axis (glucose-dependent insulinotrophic

polypeptide and glucagon-like polypeptide-1) which

augment the insulin response to oral carbohydrate

and fat, but only in the presence of a raised blood

glucose level. The cephalic phase of insulin secretion,

which occurs in anticipation of eating, is mediated by

the vagus nerve.

0007Insulin is thus the hormone of the fed state with

low circulating levels in the postabsorptive state, and

its metabolic actions (Table 3) are primarily con-

cerned with the coordinated disposal, metabolism,

and storage of the major nutrients. The major effects

are upon carbohydrate metabolism, with control of

blood glucose level being the most obvious of these.

This depends on a balance between output from the

liver and peripheral uptake, utilization and storage of

glucose; this balance is largely dependent on those

tissues which require insulin for intracellular trans-

port of glucose, particularly skeletal muscle. (See

Glucose: Maintenance of Blood Glucose Level.)

0008Since diabetes is characterized by the absence of

effective insulin activity, it is possible to view the

major metabolic features of diabetes as a reversal of

the actions in Table 3. This can be due to absolute

insulin deficiency (type 1 diabetes) or to a combin-

ation of dysfunctional insulin secretion plus insulin

resistance (type 2 diabetes).

0009Counterregulatory hormones Insulin is the only

major anabolic hormone in the body (growth hor-

mone also has anabolic properties but these are only

manifest in the presence of insulin). The action of

insulin is opposed by a number of other counterregu-

latory hormones, including glucagon, cortisol, cat-

echolamines, vasopressin, and growth hormone. The

action of these hormones is important in maintaining

fuel supplies (fatty acids and ketones as well as glu-

cose) in the fasted state. These actions are unopposed

in the absence of insulin, and in the extreme situation

of diabetic ketoacidosis they are probably responsible

for many of the metabolic abnormalities. However,

the importance of counterregulatory hormones in the

pathogenesis of diabetes is uncertain; the ratio of

tbl0002Table 2 Major physiological stimuli to insulin secretion

Metabolites Glucose

Amino acids

Ketone bodies

Gastrointestinal hormones GIP

GLP-1

Glucagon

Neural Vagal stimulation

GIP, gastric inhibitory polypeptide; GLP-1, glucagon-like polypeptide-1.

DIABETES MELLITUS/Chemical Pathology 1783

hepatic insulin to glucagon may be an important

metabolic determinant but it is interesting that dia-

betes which occurs as a result of tumors secreting an

excess of any one of the counterregulatory hormones

is usually mild, indicating the relatively greater

importance of insulin. (See Hormones: Adrenal Hor-

mones; Pituitary Hormones.)

Insulin Resistance

0010 The concept of insulin resistance has been challenged

on the grounds that it is not a discrete entity which is

always associated with pathology. This is undoubt-

edly true, and insulin sensitivity (the ‘reciprocal’ of

insulin resistance) is clearly a continuum: in a high

proportion of the nondiabetic population, the level of

insulin sensitivity can be shown to be similar to that in

people with type 2 diabetes. In the majority, normal

carbohydrate tolerance is maintained by hyper-

insulinemia, but in a proportion, insulin secretion

diminishes over time, leading to the development of

impaired glucose tolerance and, ultimately, diabetes.

The reasons for the decline in insulin secretion are

complex: ‘glucose toxicity’ (the process by which

sustained hyperglycemia results in impaired insulin

secretion) may play a role, but genetic factors are

clearly particularly important.

0011 As shown in Table 4, a number of tests have been

used to measure insulin sensitivity. The euglycemic

hyperinsulinemic clamp remains the gold-standard

method but is both labor-intensive and costly. Math-

ematical modeling of the fasting insulin-to-glucose

ratio (homeostasis model assessment: HOMA), the fre-

quently sampled intravenous glucose tolerance test and

the short insulin tolerance test are reasonable alterna-

tives. In simple terms, a high insulin-to-glucose ratio

apparently indicates insulin resistance, but there may

be difficulties in interpreting these findings. When the

high blood glucose level is taken into account, the

concomitant insulin level may, in fact, be relatively

low, although the absolute rise in insulin level with

blood glucose may be similar to that seen in nondia-

betics, suggesting a change in the ‘set’ of insulin level.

0012A further problem which complicates the interpret-

ation of insulin levels in type 2 diabetes lies in the lack

of specificity of immunoassays for insulin, nearly all

of which cross-react with either proinsulin, or insulin/

proinsulin intermediates (‘split proinsulin’), or both.

Using a range of highly specific two-point immuno-

assays for insulin, it has been shown that the level of

circulating proinsulin and insulin/proinsulin inter-

mediates (all of which are physiologically much less

active than insulin itself) is high in type 2 diabetes,

presumably resulting from dysfunctional processing

and packaging of insulin by the B cell, and that the

level of true, bioactive insulin may, in fact, be low

rather than high. To an extent, this challenges the

concept of ‘hyperinsulinemia’ in type 2 diabetes.

0013The insulin resistance of type 2 diabetes affects

both carbohydrate and fat metabolism, and is dem-

onstrable in liver, skeletal muscle, and adipose tissue.

Hepatic insulin resistance results in unsuppressed

glucose output which is the major cause of fasting

hyperglycemia, while resistance to insulin-mediated

glucose uptake in skeletal muscle is the major deter-

minant of postprandial hyperglycemia.

0014The adipocyte is more sensitive to the action of

insulin than either muscle or liver, but the dose–

response curve is shifted to the right in type 2

tbl0004Table 4 Tests used to measure insulin sensitivity

Fasting insulin-to-glucose ratio (IGR)

Mathematical modeling of IGR (homeostasis model assessment,

HOMA)

Frequently sampled intravenous glucose tolerance test (FSIVGTT)

Short insulin tolerance test (ITT)

Euglycemic hyperinsulinemic clamp

Adapted from Hermans MP, Levy JC, Morris RJ and Turner RC (1999)

Comparison of insulin sensitivity tests across a range of glucose tolerance

from normal to diabetes. Diabetologia 42: 678–687.

tbl0003 Table 3 The metabolic actions of insulin

Metabolism Action Site (tissue) of action

Carbohydrate metabolism Stimulates glucose transport,

glucose phosphorylation,

glycolysis, and

glycogen synthesis

M,F

M,F,L

M,L

Inhibits gluconeogenesis and

glycogen breakdown

L,K

M,L

Fat metabolism Stimulates fatty acid synthesis and

triglyceride synthesis

L,F

Inhibits lipolysis L,F

Protein metabolism Stimulates amino acid transport and

protein synthesis

L,M

Inhibits protein breakdown L,M

M, muscle; F, fat; L, liver; K, kidney.

1784 DIABETES MELLITUS/Chemical Pathology

diabetes, and adipose tissue lipolysis (particularly

central adipose tissue) is increased in the fasting

state and incompletely suppressed postprandially. As

a result, plasma free (non esterified) fatty acid levels

are characteristically elevated, but, in type 2 diabetes,

since insulin is always present, lipolysis is never com-

pletely uninhibited. As a result, ketosis is not a feature

of type 2 diabetes. In contrast, in untreated type

1 diabetes resulting from absolute insulin deficiency,

not only is lipolysis uninhibited, but reesterification

of free fatty acids to form triglycerides is impaired

because of the reduction in supplies of a-glyceropho-

sphate from glycolysis. In the liver, glycolysis is not

insulin-dependent and triglyceride synthesis is unim-

paired and may even be increased because of an in-

creased supply of free fatty acids. (See Fatty Acids:

Metabolism.)

Hyperglycemia and Glycosuria

0015 Blood glucose concentration is normally carefully

controlled and maintained between fairly narrow

limits as a result of a balance between hepatic output

and peripheral uptake of glucose. Even following

carbohydrate feeding, around 60% of absorbed glu-

cose is extracted during the first pass through the liver

under the influence of insulin, such that the post-

prandial rise in blood glucose is normally modest.

0016 Glucose is freely filtered by the glomerulus and

reabsorbed by an active transport system in the prox-

imal tubule. Under normal circumstances, the urine

contains virtually no glucose as the capacity of the

renal tubules to reabsorb glucose exceeds the filtered

load. However, as the blood glucose rises, there

comes a point at which the filtered load of glucose

in the renal tubule exceeds the absorptive capacity,

and glucose appears in the urine; this is the renal

threshold for glucose. The normal renal threshold is

between 6 and 7 mmol l

1

. It is lower in those indi-

viduals with idiopathic renal glycosuria and during

pregnancy. The presence of glucose in the urine exerts

an osmotic effect in the distal and collecting tubules

of the kidney, countering physiological mechanisms

for urine concentration, resulting in an increase in

urine volume (polyuria). The resultant loss of water

from the body (plus the direct effect of hyperglycemia

on osmoreceptors in the hypothalamus) results in

polydipsia (excess thirst) and, if fluid intake is

inadequate, dehydration.

Clinical Features and Symptoms

Type 1 Diabetes

0017 The classical features of type 1 diabetes are weight

loss, polydipsia, and polyuria. This is the direct result

of the osmotic diuresis which results from hypergly-

cemia; the weight loss results from dehydration plus

loss of both fat and muscle due to the catabolism

which occurs in the absence of insulin. In addition

to these specific symptoms, there is frequently a

history of increasingly severe, nonspecific symptoms

such as tiredness, lack of energy, muscle cramps, and

general ill health, often over a period of some months,

which may only be recognized retrospectively after

the development of full-blown diabetes. In children

particularly, the onset may be more rapid, with

ketoacidosis supervening before the diagnosis is

established.

Type 2 Diabetes

0018In contrast to type 1, the symptoms of type 2 diabetes

are very insidious, frequently very mild, and often

absent altogether; up to 50% of type 2 diabetes in

the general population may be undiagnosed. Most

patients are overweight. Indeed, older patients pre-

senting with weight loss are likely to have type 1 dia-

betes and to require treatment with insulin from the

outset. Polyuria and polydipsia may be noticed, and a

history of urinary tract infection and vulval candidia-

sis can frequently be elicited. Ketoacidosis does not

occur but occasional patients present with hyper-

osmolar, nonketotic coma.

0019Complications such as neuropathy, arterial disease,

and retinopathy are frequently present at diagnosis

and may be a presenting feature. In many patients,

however, the condition remains asymptomatic and is

only detected on routine screening.

Coma in Diabetes

0020Three types of coma occur in diabetes:

0021Hypoglycemic coma.

0022Hyperglycemic coma, either ketoacidotic coma

(diabetic ketoacidosis: DKA), or hyperosmolar,

nonketotic coma.

Hypoglycemic Coma

0023Hypoglycemia occurs predominantly in patients

taking insulin but also in patients taking sulfonyl-

ureas, particularly long-acting preparations such as

chlorpropamide and glibenclamide. The factors

which commonly predispose to hypoglycemia, either

alone or in combination, are:

0024too much insulin or sulfonylurea

0025too little food or delayed meal

0026too much physical exercise.

In addition to simple overdosing with insulin, vari-

ations in absorption from different injection sites, or

DIABETES MELLITUS/Chemical Pathology 1785