Carbohydrate metabolism


Carbohydrate metabolism


Carbohydrate metabolism


Carbohydrate metabolism

3.1 Diabetes mellitus

Lothar Thomas

3.1.1 Diabetes mellitus

Diabetes mellitus is a group of metabolic diseases characterized by hyperglycemia resulting from defects in insulin secretion, insulin action, or both /1/. Long-term chronic hyperglycemia includes dysfunction and damage of organs such as retinopathy, nephropathy, peripheral neuropathy, atherosclerotic cardiovascular, peripheral arterial, and cerebrovascular disease. Diabetes may be diagnosed based on the following criteria:

  • Plasma glucose or blood glucose (fasting glucose)
  • 2-h plasma glucose or blood glucose value during a 75-g oral glucose tolerance test
  • HbA1c criteria.

Generally, all tests are equally appropriate for diagnostic testing, however the tests do not necessarily detect diabetes in the same individuals. Unless there is a clear clinical diagnosis (e.g., patient in a hyperglycemic crisis or with classic symptoms of hyperglycemia and a random plasma glucose > 200 mg/dL; 11.1 mmol/L), a second test is required for confirmation. If a patient has discordant results from two different tests, than the test result that is above the diagnostic cut point should be repeated /2/. Epidemiology of diabetes mellitus

According to estimates /1/, the world prevalence of diabetes among adults aged 20–79 is 6.4%. It varies between continents: North America 10.2%, Middle East (EMME) 9.3%, South-East Asia 7.6%, Europe 6.9%, South America 6.6%, and Africa 3.8%. More than 90% have type 2 diabetes. This type, formerly known as adult-onset diabetes, not only affects adults, but increasingly also children and adolescents. The most common risk factors for type 2 are overweight and lack of physical activity. Estimated incidence rates from the United States of type 1 diabetes increased by 1.4% annually and of type 2 diabetes among non Hispanic whites by 0,6% annually in the 2002–2012 period. The incidences increased particularly among youths of minority racial and ethnic groups /4/.

According to estimates of the Organization for Economic Cooperation and Development (OECD), 60% of men and 45% of women in Germany are overweight, 16% are obese, and diabetes associated mortality was 175,000 cases in 2010. The prevalence of overweight in adolescents is 15–20%, with 8% being obese. 85% of overweight adolescents will remain overweight as adults and a significant number of them will develop diabetes at an early age. Secondary diseases of diabetes include:

  • Cardiovascular disease; three thirds will die of it
  • Stroke; occurs four times more frequently in diabetics than in the general population
  • Dementia; the risk is 3-fold higher, or 11-fold higher in the presence of hypertension
  • The incidence of depression and Parkinson’s disease is twice as high in patients with diabetes than in the general population. Classification of diabetes mellitus

Diabetes is classified into the following general categories/23/:

  • Type 1 diabetes due to autoimmune β-cell destruction, usually leading to absolute insulin deficiency
  • Type 2 diabetes due to progressive loss of β-cell insulin secretion frequently on the background of insulin resistance
  • Gestational diabetes mellitus. This diabetes type is diagnosed in the second or third trimester of pregnancy that was not clearly overt diabetes prior to gestation.
  • Specific types of diabetes due to other causes (Tab. 3.1-1: Classification of diabetes mellitus).

Classification is important for determining therapy, but some individuals cannot be clearly classified as having type 1 diabetes or type 2 diabetes. Both types are heterogenous diseases in which clinical presentation and disease may vary considerably. Once hyperglycemia occurs, patients with all forms of diabetes are at risk for developing the same chronic complications, although rates of progression may differ. The traditional paradigms of type 1 diabetes occurring only in children and type 2 diabetes occurring only in adults are no longer accurate, as both diseases occur in both age groups /2/. Categories of increased risk for diabetes (prediabetes)

Prediabetes is the term used for individuals whose glucose levels do not meet the criteria for diabetes but are too high to be considered normal. Screening for prediabetes and risk for future diabetes in asymptomatic people should be considered in adults of any age who are overweight or obese (BMI ≥ 25 kg/m2 or ≥ 23 kg/m2 in Asian Americans /2/. Pathogenic processes involved in the development of diabetes mellitus

Pathogenic processes in diabetes mellitus vary and range from autoimmune destruction of the β-cells of the pancreas with insulin deficiency to abnormalities that result in resistance of the tissues to insulin action. The basis of the abnormalities in carbohydrate, fat, and protein metabolism in diabetes mellitus is deficient action of insulin on target organs. Deficient insulin action can result from inadequate insulin secretion or diminished biological tissue responses to insulin (insulin resistance). Impairment of insulin secretion and diminished tissue response often coexist in the same patient, and it is unclear which of the two processes is the primary cause of the hyperglycemia.

Insulin resistance is due to a decrease in insulin sensitivity of the tissues (insulin resistance = 1/insulin sensitivity). Insulin sensitivity is the ability of insulin to lower plasma glucose concentration by reducing hepatic glucose production and stimulating glucose uptake in insulin-sensitive tissues, in particular skeletal muscle and adipose tissue.

The cell damage in diabetes is due to hyperglycemia. The damaging effect is caused by repeated rapid change in the cell metabolism and the persistent hyperglycemia. The degree of hyperglycemia in diabetics changes over time, depending on the disease process. In addition to the accelerating factors hyperlipidemia and hypertension, the hyperglycemia-induced increased production of super oxides causes induction of DNA strand breaks. Symptoms and long-term complications of diabetes mellitus

Symptoms of marked hyperglycemia

Polyuria, polydipsia, weight loss, sometimes with polyphagia, and blurred vision. Acute, life-threatening consequences of uncontrolled diabetes are hyperglycemia with ketoacidosis, and hyper osmolar non ketotic syndrome. Chronic hyperglycemia may also be accompanied by growth impairment in children and increased susceptibility to infections.

Long-term complications of hyperglycemia result from micro- and macro vascular injury:

  • Microvascular damage includes retinopathy with loss of vision, nephropathy leading to renal failure, peripheral neuropathy with risk of foot ulcers, autonomic neuropathy causing gastrointestinal, genitourinary and cardiovascular symptoms, and sexual dysfunction
  • Macro vascular complications are cardiovascular diseases (myocardial infarction), peripheral vascular diseases (circulation disorder such as of the leg arteries), and cerebrovascular diseases (stroke). Accelerating factors are hypertension and hyperlipidemia /3/. Criteria for testing for diabetes or prediabetes in asymptomatic adults

1. Testing should be considered in all adults who are overweight (BMI ≥ 25 kg/m2) and who have one or more additional risk factors /3/:

  • First-degree relative with diabetes mellitus
  • High-risk race/ethnicity (e.g., Native American, African American, Latino, Asian American, Pacific Islander)
  • History of cardiovascular disease
  • Hypertension (blood pressure ≥ 140/90 mmHg) or on therapy for hypertension
  • HDL cholesterol level < 35 mg/dL (0.90 mmol/L) and/or triglycerides > 250 mg/dL (2.82 mmol/L)
  • Women with polycystic ovary syndrome
  • Physical inactivity
  • Other conditions associated with insulin resistance (severe obesity, acanthosis nigricans)

2. HbA1c ≥ 5.7% (39 mmol/mol), impaired glucose tolerance or impaired fasting glucose should be tested yearly.

3. Women who are diagnosed with gestational diabetes mellitus should have lifelong testing at least every 3 years.

4. For all other patients, testing should begin at age 45 years.

5. If results are normal, testing should be repeated at a minimum of 3-year intervals, with consideration of more frequent testing depending on initial results and risk status. Testing for type 2 diabetes in asymptomatic children

Overweight (BMI > 85th percentile for age and sex, weight for height > 85th percentile or weight > 120% of ideal for height). Plus any two of the following risk factors /3/:

  • Family history of type 2 diabetes in first- or second-degree relative
  • Race and ethnicity (Native American, African American, Latino, Asian American, Pacific Islander)
  • Signs of insulin resistance or conditions associated with insulin resistance (acanthosis nigricans, hypertension, dyslipidemia, polycystic ovary syndrome, or birth weight small for gestational age birth weight)
  • Maternal history of diabetes or gestational diabetes during the child’s gestation
  • Age of initiation: 10 years at onset of puberty, if puberty occurs at younger age.

Frequency: every 3 years. Diagnosis of prediabetes and diabetes mellitus

Criteria for testing for type 1 and type 2 diabetes mellitus and prediabetes are listed in Tab. 3.1-2 – Diagnosis of prediabetes and diabetes mellitus based on ADA criteria. Etiologic types and stages of diabetes

In the period from 2010 to 2030, the number of adult diabetics is predicted to increase by 69% in the developing countries and by 20% in the developed countries /1/. By the year 2030, over 75% of people with diabetes will reside in developing countries and will be in the age range of 45–64 years.

By contrast, in the developed countries, the majority of diabetics will be above 65 years of age. There will be more women with diabetes than men, especially in the developed countries.

Diabetes type I: 5–10% of diabetics are type I. The incidence differs greatly between countries (Japan: 1 in 100,000 children and adolescents per year, Germany: 7, USA: 7, Finland: 35). The prevalence is 0.1–0.3% in Europe and 0.4% in the USA. There are approximately 2 million type 1 diabetics in Europe and North America. Globally, the incidence increases by about 3% per year, partly due to the fact that the average age of onset is decreasing. Type 1 is diagnosed in children, adolescents and young adults. It does not occur until 6 months of age, after which the incidence increases continuously with age, peaks in puberty, and then declines to half its peak in adulthood /5/.

Diabetes type 2: approximately 95% of diabetics in Europe and North America have type 2 diabetes. The prevalence in Europe and in U.S. Caucasians is 6%, in Blacks it is 10%. In Germany, the prevalence in adults aged 18–79 is 7.2%. From the age of 50 it rises continuously to 20% in adults aged 70–79. 2.1% of adults have undiagnosed diabetes /6/. A 50-year-old diabetic can expect to live 5.8 years less than a nondiabetic person of the same age.

The number of children and adolescents with type 2 diabetes is steadily increasing. In Pennsylvania (USA), the incidence of type 2 diabetes in white adolescents aged 15–19 is 11.2 per 100,000 per year. In Central Europe, 0.5–1% of diabetic children aged 0–16 have type 2 diabetes.

Prediabetes: depending on the test used for diagnosis, one in four US Americans over the age of 20 have prediabetes.

3.1.2 Type 1 diabetes (T1D)

This type of diabetes was previously known as insulin-dependent diabetes or juvenile diabetes and accounts for 5–10% of diabetes. Clinically manifest T1D is preceded by a preclinical period of varying length. During this phase, autoimmune processes in the pancreatic islets reduce the β-cells to such an extent that the blood glucose concentration can no longer be maintained within the physiological range. Type 1 has two subtypes (Tab. 3.1-1 – Classification of diabetes mellitus):

  • Type 1A, the immune-mediated form
  • Type 1B, the idiopathic form.

Type 1A

The auto-immune form of diabetes is characterized by immune-mediated destruction of β-cells, leading to absolute insulin deficiency /1/. The rate of β-cell destruction is variable, being rapid in some individuals and slow in others. In some patients, especially children and adolescents, the disease may first manifest as ketoacidosis. Others have modest fasting hyperglycemia which can rapidly progress to severe hyperglycemia with ketoacidosis in the presence of an infection or stress. Still others, in particular adults, may retain residual β-cell function sufficient to prevent ketoacidosis for many years and will not become dependent on insulin until later. At this latter stage of the disease, there is no more detectable insulin or C-peptide secretion. Although type 1A diabetes typically develops in children and adolescents, it can also occur later in life. The incidence is about the same for both sexes.

Markers of the immune-mediated destruction of β-cells include the following islet cell antigen autoantibodies: anti-insulin, anti-glutamic acid decarboxylase (GAD65), anti-tyrosine phosphatase IA-2 and IA-2 β. At least one but usually several of these antibodies are present in 85–90% of patients when fasting hyperglycemia is initially detected.

Type 1A has a strong HLA association and shows familial aggregation. It is also associated with other autoimmune diseases such as Graves’ disease, Hashimoto’s thyroiditis, Addison’s disease, celiac disease, autoimmune hepatitis, myasthenia gravis, pernicious anemia or vitiligo in the patient him/herself or a close family member.

Type 1B

This type of diabetes has no known etiology. Multiple forms of β-cell dysfunction are associated with this type of diabetes. Only a small portion of type 1 diabetics fall into this category. Type 1B lacks evidence of an autoimmune process and has no HLA association, but it is strongly inherited. Most patients are of African or Asian ancestry. They suffer from episodic ketoacidosis and exhibit varying degrees of insulin deficiency between episodes. Type 1B belongs to the ketosis-prone diabetes group /7/. Staging of type 1 diabetes /2/

Stage 1 characteristics

  • Autoimmunity, normoglycemia, pre symptomatic

Stage 1 diagnostic criteria

  • Multiple autoantibodies, no impaired fasting glucose (IFG), no impaired glucose tolerance (IGT)

Stage 2 characteristics

  • Autoimmunity, dysglycemia, pre symptomatic

Stage 2 diagnostic criteria

  • Multiple autoantibodies, dysglycemia (IFG and/or IGT): FPG 100–125 mg/dL (5.6–6.9 mmol/L), 2-h plasma glucose 140–199 mg/dL (7.8–11.0 mmol/L), HbA1C 5.7–6.4% (39–47 mmol/mol) or ≥ 10% increase in HBA1C

Stage 3 characteristics

  • New onset hyperglycemia, symptomatic

Stage 3 diagnostic criteria

  • Clinical symptoms, diabetes by standard criteria Pathophysiology of type 1A diabetes

Type 1A is a multifactorial and polygenic disease. Polygenic means that multiple genetic mutations are required for the disease to manifest. Factors of the immune-mediated destruction of β-cells are /8/:

  • Genetic predisposition for this disease.
  • Environmental factors that trigger the immunological process
  • Expression and activation of target antigens to support the immune process. Genetic predisposition

Type 1A is thought to be caused by an interaction between different genetic and environmental factors. For example, the concordance rate among monozygotic twins is less than 100%, and even though type 1A aggregates in the family, there is no clear mode of inheritance /89/. There is a clear association between the genes on chromosome 6p21 and type 1A. These are class II genes which encode HLA-DR and HLA-DQ markers. Therefore, one of the key focuses in predicting the genetic risk of type 1A is the genotyping of the HLA-DR and HLA-DQ loci. For example, children who carry both high-risk haplotypes (DR3-DQ2 and DR4-DQ8) have a 1 : 20 risk of developing type 1A by the age of 15. If one monozygotic twin is diabetic and the other carries both haplotypes, the risk of developing diabetes is as high as 55%. About 35% of type 1A diabetics among the white population in the USA are DR3-DR4 heterozygous compared to 2.4% of the general population. It is thus possible to identify high-risk groups in combination with autoantibody testing.

A number of non-HLA-associated gene loci that are associated with type 1A were identified. The odds ratio of the loci for the association with type 1A is only 1.2–1.5, with the exception of that of the loci INS and PTPN22, which is 2–2.5. INS is the gene for insulin and the primary auto antigen in type 1A. CTLA4 plays a role in the T-cell development and auto antigen recognition, and PTPN22 encodes the tyrosine phosphatase of T-cells and is involved in T-cell receptor signaling.

It is thought that 1–5% of the Caucasian population carry genes associated with a high risk of type 1A.

The following proportions are reported for the risk of type 1A /210/:

  • Total population 0.4%
  • Familial aggregation, defined as the risk in siblings as compared to the total population (6%/0.4%; risk is about 15 times higher)
  • HLA DR3-DR4 heterozygous individuals 7%
  • First-degree relatives (siblings or parents with type 1A diabetes) 6%, child born to a mother with type 1A diabetes 1.3–4%, child born to a father with type 1A diabetes 6–9%
  • First-degree relatives (siblings or parents with type 1A) with HLA DR3–DR4 heterozygosity 20–30%
  • Monozygotic twin of a type 1A diabetic 30–70%.

However, 85% of cases of type 1A diabetes occur sporadically i.e., in the absence of a first-degree relative with immune-mediated diabetes. Therefore the determination of genetic markers is currently considered of little value in the diagnosis of type 1A. Environmental factors

Evidence of environmental factors contributing to the development of type 1A is based on the observation of seasonal and geographical variations in the incidence of the disease. For example, the incidence is lower in infants who received breast milk compared to those who were fed formula.

Viral infections are considered to be an important trigger of type 1A. The virus either directly destroys the β-cells through its cytopathogenic effect, or it triggers a chronic inflammatory process during which the β-cells are damaged. Finally, the destruction of islet cells can also result from a similar antigen structure of viral and β-cells (molecular mimicry).

Viral infection: molecular mimicry is reported to be a main cause of virus-induced autoimmune-mediated diabetes. Examples of viral infections that can trigger type 1A /11/:

  • Congenital Rubella virus infection. More than 90% of genetically predisposed children (HLA DR3 or DR4 positive) with a congenital rubella infection develop diabetes
  • Severe Coxsackie B virus infection. There are cross-reacting antigens between Coxsackie B4 virus proteins and the β-cell auto antigen GAD65. Target antigens, expression and activation

The immune system begins to produce antibodies to β-cell antigens long before clinical symptoms of type 1A diabetes appear. These autoantibodies, also known as islet cell antibodies (ICA), are directed against sequestrated antigens of the β-cells and sustain an immune response which ultimately leads to the destruction of the β-cells. Clinical symptoms of diabetes appear when about 80% of the β-cells are dysfunctional. Since ICA are stimulated by antigens, their titer decreases over time. There is as yet no clear evidence that the ICA play a causal role in type 1A. They are, however, disease markers that are produced in the destruction of β-cells as an immune-mediated response to the release of β-cell proteins /9/.

The inflammatory injury of the pancreas that occurs in type 1A is known as insulitis. The early stage of inflammation is characterized by mononuclear cell infiltration of the acini (peri-insulitis) which surround the islets of Langerhans. Once these cells penetrate the islets, their destruction begins. The preclinical and clinical phases associated with the loss of islet cell mass are shown in Fig. 3.1-1 – Development of type 1A diabetes.

Criteria of type 1A diabetes and relevant screening tests are shown in Tab. 3.1-1 – Classification of diabetes and Tab. 3.1-2 – Diagnosis of prediabetes and diabetes based on ADA criteria.

3.1.3 Categories with increased risk for diabetes (prediabetes)

Prediabetes is a condition of impaired glucose metabolism rather than an intermediate stage or risk factor for predicting /12/:

  • The development of diabetes mellitus (increased risk for diabetes)
  • The increased risk of cardiovascular or microvascular disease.

Prediabetes can be diagnosed on the basis of impaired fasting glucose (IFG), impaired glucose tolerance (IGT) or an elevated HbA1c level (Tab. 3.1-2 – Diagnosis of prediabetes and diabetes mellitus based on ADA criteria). The progression from prediabetes to diabetes (up to 70%) can take years, but can also occur rapidly or not at all. The incidence is highest in individuals with combined IFG and impaired glucose tolerance and the incidence is 5–10% per year. The main causes of prediabetes are increasing insulin resistance and changed lifestyle factors. Insulin sensitivity/insulin resistance

The pathophysiology of prediabetes is characterized by alteration in the body cells’ sensitivity to insulin and a change in the insulin secretion of the β-cells of the pancreas /12/. The cells, in particular skeletal muscle, hepatocyte and adipose tissue cells, become increasingly insulin resistant, and insulin-mediated glucose uptake is reduced. The insulin sensitivity of the cells is reversely related to the glucose concentration and directly related to the insulin resistance. As fasting glucose increases from 70 mg/dL (3.9 mmol/L) to 125 mg/dL (6.9 mmol/L), insulin sensitivity decreases by a factor of 3. Individuals with isolated elevation of fasting glucose still have approximately 80% insulin sensitivity while those with elevated fasting glucose and impaired glucose tolerance have lost 80% of their insulin sensitivity.

Most individuals respond to insulin resistance with an increase in β-cell mass and an adaptive increase in insulin secretion in order to maintain a normal glucose concentration. In the presence of hyperinsulinism, normal or nearly normal glucose tolerance is maintained for years or decades, thus preventing diabetic hyperglycemia and an increase in free fatty acids.

The main effects of insulin resistance, such as diminished glucose uptake by cells, deficient inhibition of endogenous gluconeogenesis and a reduced number of glucose transporters in the fat and muscle cell membranes, also occur in the presence of hyperinsulinism. Individuals with insulin resistance thus require a two-fold higher insulin concentration than healthy individuals for half-maximal suppression of gluconeogenesis and glucose uptake by the cells.

The relationship between insulin resistance and the amount of insulin required to overcome the resistance is hyperbolic (Fig. 3.1-2 – Relationship between maximum insulin response to glucose and insulin sensitivity). This is likely one of the causes why individuals who are genetically predisposed for insulin resistance (e.g., first-degree relatives of a T2D), remain glucose tolerant through many years. In the prediabetic years insulin resistance is low and requires little additional secretion of insulin. Only when the insulin response to glucose reaches the steep slope of the insulin response/insulin sensitivity curve, is compensation no longer possible. This usually occurs when these individuals become older, physically less active and overweight, but is not the case in all older and overweight individuals, since 80% of them remain glucose tolerant. Therefore, patients who develop T2D must additionally have genetic β-cell destruction causing changes in insulin secretion as early as in the prediabetic phase. The relationship between maximum insulin response and the plasma glucose concentration is shown in Fig. 3.1-3 – Relationship between maximum insulin response to arginine stimulation and plasma glucose. Lifestyle factors

An atypical lifestyle plays a role in the development of prediabetes and T2D. It leads to stress of the endoplasmic reticulum, dysfunction of the mitochondria, excessive ectopic fat deposition, change in innate immunity, and insulin resistance. As a result, affected individuals develop local or systemic low-grade inflammation to which their immune system cannot respond adequately /14/. To prevent and delay the development of T2D, it is recommended that these individuals be physically active for at least 150 min./week and strive for 7% weight loss. For those with a BMI > 35 kg/m2 metformin therapy may be considered. It is thought that the excessive production of cytokines such as resistin, visfatin, angiotensinogen, PAI-1 and TNF-α by adipose tissue as well as the diminished synthesis of positive adipocytokines such as adinopectin contribute to the inflammation. The progression from prediabetes to T2D is shown in Fig. 3.1-4 – Development of type 2 diabetes in three stages. Risk of progression from prediabetes to diabetes

Studies on the prevention of prediabetes and T2D have shown that the progression from prediabetes to T2D can be prevented or delayed through lifestyle intervention (change of diet, weight loss, increased physical activity), pharmacotherapy (e.g., metformin) or, in the case of heavy obesity, through bariatric surgery. In addition to the components of the metabolic syndrome (visceral obesity, hypertension, dyslipidemia), biomarkers also are important indicators for evaluating the risk for diabetes (see also Section 2.2 – Metabolic syndrome). Risk scores are used in particular to identify high-risk patients in order to alleviate secondary complications.

Two of these risk scores for diabetes are:

  • The diabetes risk test (DRT; webtool, www.dife.de), the German adaptation of the Finnish score (FINDRISK). It applies for individuals aged 35 and older and determines the overall risk of developing T2D within the next 5 years /16/. The overall risk considers age, anthropometric data (waist circumference, height), hypertension as well as nutrition and lifestyle related variables (frequency of consuming wholemeal bread, red meat, coffee, alcohol, smoking, activity profile). Based on a cutoff score of ≥ 49, the test identifies the T2D cases expected in the normal population over the next 5 years with a diagnostic sensitivity of 85% and a specificity of 68%. By additionally performing biomarkers such as plasma glucose, HbA1c, HDL cholesterol, triglycerides, GGT and ALT, the diagnostic sensitivity and specificity are improved. A fasting plasma glucose ≥ 100 mg/dL (5.6 mmol/L) alone increases the diagnostic specificity to 85%.
  • The scores derived from the US Atherosclerosis Risk in Communities (ARIC) study. The base score contains the parameters hip circumference, maternal diabetes, hypertension, paternal diabetes, microsomia, black race, age > 55, overweight, increased heart rate, and smoking. The incidence of type 2 diabetes in the highest quintile is 33% for the subsequent 10 years. Addition of the biomarkers glucose, triglycerides, HDL cholesterol and uric acid to the score increases the incidence to 46.1%, if plasma glucose is ≥ 106 mg/dL (5.88 mmol/L ), triglycerides are ≥ 179 mg/dL (2.02 mmol/L ), HDL cholesterol is < 40 mg/dL (1.02 mmol/L ) and uric acid ≥ 7.8 mg/dL (464 μmol/L ) in men /17/. Postprandial and post absorptive glucose

The behavior of plasma glucose and insulin in the post absorptive (fasting) state (fasting plasma glucose, FPG) and postprandial following glucose load in the oral glucose tolerance test (oGTT) are important criteria for assessing insulin secretion (fasting level) and insulin resistance (2-h level). Glucose ingestion leads to increased insulin secretion. Normally, plasma glucose is taken up by the peripheral insulin-sensitive tissues in the postprandial state. If there is insulin resistance, glucose clearance is delayed, resulting in hyperglycemic glucose levels. Combined FPG and oGTT testing provides information about insulin secretion via the FPG level and information about insulin resistance via the 2-h glucose level from the oGTT.

This can produce the following results, which are indicative of the presence of prediabetes:

  • Elevated FPG with normal 2-h glucose level; a condition termed impaired fasting glucose (IFG).
  • Normal FPG with elevated 2-h glucose level; a condition termed impaired glucose tolerance (IGT).
  • Combination of IFG and IGT without reaching the diagnostic thresholds for type 2 diabetes; a condition termed IFG/IGT.

Tab. 3.1-3 – Post absorptive and postprandial glucose in non-diabetics and diabetics shows the post absorptive and postprandial behavior of glucose in healthy individuals and diabetics. Criteria and tests for prediabetes screening in asymptomatic individuals are shown in Tab. 3.1-2 – Diagnosis of prediabetes and diabetes mellitus based on ADA criteria.

3.1.4 Type 2 (T2D) diabetes

In patients with T2D the paths to β-cell demise and dysfunction are less well defined, but deficient β-cell insulin secretion, frequently in the setting of insulin resistance, appears to be the common denominator /2/. T2D is characterized by chronic hyperglycemia and often remains undiagnosed for years, because glycemia increases gradually and does not cause any symptoms in the early stages. These patients are nevertheless at increased risk of developing micro- and macro vascular complications. T2D develops as follows (Fig. 3.1-4 – Development of type 2 diabetes in three stages):

  • It starts with a genetic predisposition
  • As insulin resistance develops, glucose tolerance becomes impaired, a condition called prediabetes
  • Subsequent to insulin resistance there is reduced β-cell function, resulting in the development of T2D.

Pathophysiological causes /23/:

  • Insulin resistance (see Section 3.1.3 – Categories with increased risk for diabetes (prediabetes))
  • A relative, and later absolute, deficiency in insulin secretion. The patient usually does not depend on insulin for survival. As diabetes progresses, insulin resistance is no longer compensated by insulin secretion. In this situation, which occurs mainly in older people, in individuals who are obese or physically inactive, in patients with hyperlipidemia or hypertension and women with a history of gestational diabetes, it may be possible to alleviate insulin resistance by weight loss and treatment of hyperglycemia.

T2D is a multifactorial disease in which genetic and epigenetic factors play an important role. Genetic factors

The significance of genetic factors is reflected by the fact that ethnic populations, such as the Pima Indians, have a diabetes prevalence of up to 21%. The strong genetic basis of type 2 diabetes has also been demonstrated by tests on twins which show that the concordance of T2D in monozygotic twins is 70% compared with only up to 30% in dizygotic twins. T2D is a polygenic disease which is caused by the concurrent occurrence of many DNA sequence variations in different genes /21/. Each of these variations alone has only a moderate effect on the relevant function or expression of the gene, but in combination they lead to higher sensitivity to adverse environmental factors. Currently there are over 60 gene loci with variants that influence the risk of developing type 2 diabetes. Usually these are single nucleotide substitutions of one base for another (single nucleotide polymorphisms, SNPs) which are of different relevance for the risk of diabetes for the different ethnic groups. Relevant SNPs are known in the following loci: PPARG, KCNJ11, TCF7L2, HHEX, CDKALI, CDKN2A/B, IGF2BP2, SCL30A8 and WFS1. However, the associations are weak, and each variant increases the risk for diabetes only by a factor of 1.05–1.4. Even though these gene variants contribute to the risk of diabetes, they do not allow a better prediction of the risk than the clinical risk factors. Epigenetic factors

The main epigenetic factors are older age, obesity, and lifestyle factors. Obesity is caused by excess consumption of calories, in particular unsaturated fats, sugary or starchy foods, reduced consumption of dietary fiber, a sedentary lifestyle, and reduced physical activity /13/. Although obesity is an important factor, 10% of type 2 diabetics are of normal weight, and not every obese person will develop diabetes. Psychosocial factors such as sleep deprivation and depression are also important factors while smoking and infections play a minor role.

Symptoms: because T2D has few symptoms, its clinical diagnosis is delayed. Only marked hyperglycemia will lead to symptoms such as fatigue, weakness, polyuria and polydipsia. Ketoacidosis is rare and only occurs when there is an infection, another disease or a stress situation /22/. The results of the United Kingdom Prospective Diabetes Study allow the conclusion that patients with type 2 diabetes still have 50% of normal β-cell function at diagnosis, but only 25% 6 years later. Extrapolation back to 100% β-cell function allows the conclusion that the decline of insulin secretion began 10–12 years before clinical symptoms appeared /23/. Pathophysiology of type 2 diabetes

T2D is characterized by insulin resistance, impaired insulin secretion and reduced β-cell function.

Insulin resistance: also refer to prediabetes and Section 2.2 – Metabolic syndrome/17/.

β-cell mass and T2D /20/: the cause of the diminished β-cell function in T2D is unknown. One significant factor is the reduction in the number of β-cells: in obese individuals with elevated fasting glucose levels, the number of β-cells is reduced by 50% compared to healthy individuals, and the United Kingdom Prospective Diabetes study showed that insulin secretion in type 2 diabetics is also reduced to 50% at diagnosis. The diminished β-cell function leads to the following disorders in T2D:

The apoptosis of β-cells in T2D is reported to be attributable to the intracellular oligomerization of islet amyloid polypeptide (IAPP). IAPP is co-expressed and co-secreted with insulin by the islet cells. IAPP inhibits insulin secretion by exerting a direct paracrine effect on these cells. One assumption is that, in T2D, IAPP is misfolded in the cells, causing it to form cytotoxic aggregates and induce the apoptosis of the β-cells /25/.

3.1.5 Other specific types of diabetes with known causes Maturity-onset diabetes of the young (MODY)

In addition to TD1 and T2D, MODY comprises a small group of non-insulin-dependent diabetics, in whom the disease manifests in childhood and young adulthood. Patients have highly penetrating autosomal dominant mutations in a single gene (monogenic diabetes) which lead to dysfunction of the β-cells /2627/. There are different phenotypes with different characteristics such as age of occurrence, severity of hyperglycemia, response to treatment, secondary diseases, and extra pancreatic diseases. The prevalence of MODY is estimated to be 0.6–2% of all diabetes cases, although in reality it is probably higher since MODY is often incorrectly classified as TD1 or TD2. Heterozygous mutations in the genes GCK and HNF1A/4A account for up to 80% of MODY cases. GCK encodes the intracellular enzyme glucokinase, which acts as a glucose sensor in the pancreatic β-cells. Mutations in GCK lead to mild, often asymptomatic hyperglycemia whereas mutations in the genes encoding the transcription factors hepatocyte nuclear factor-1α and -4α cause progressive insulin deficiency with hyperglycemia which can lead to vascular complications. The less common type of MODY, which results from mutations of the transcription factor gene HNF1B, is associated with extra pancreatic manifestations such as cystic kidney disease. The different types of MODY are listed in: Differential diagnosis of T1D, T2D and MODY

Differentiation of MODY from T1D: MODY develops slowly, with mild hyperglycemia and an increase in insulin levels in response to glucose load, without progression to ketoacidosis and negative pancreatic autoantibodies. Type 1 manifests in children and adolescents of lean habitus and usually occurs acutely with ketoacidosis. A differential diagnosis is shown in Fig. 3.1-6 – Differential diagnosis of MODY, type 1 and type 2 diabetes, but it is not always straightforward. Refer also to Fig. 3.1-5 – Important organs and tissues involved in glucose metabolism.

Differentiation of MODY from T2D: T2D often manifests at a later age (over 40 years) than MODY and there are co-morbidities or overweight, which is less frequently the case in MODY. MODY is an autosomal dominant disease which is inherited through 3 generations. Patients with MODY are generally lean rather than overweight. Latent autoimmune diabetes in adults (LADA)

The pathophysiology of LADA is far less understood than that of T1D and T2D. However, its clinical manifestation contains characteristics of both these types of diabetes. LADA thus presents as follows:

  • Clinically: as T2D, with onset in adulthood and without insulin dependency at diagnosis
  • Diagnostically: like T1D due to positive result for islet cell antibodies
  • Genetically: through association with the gene locus TCF7L2 as well as MHC /28/.

8–10% of diagnosed T2D cases are likely to be LADA. The definition of LADA as a separate subgroup is based on the understanding that this type of diabetes, which presents in adulthood and with autoantibodies, is non-insulin-dependent at diagnosis and its progression to insulin dependency is slower than in T1D. The “adult age of onset” varies between 25 and 40 years.

The presence of glutamic acid decarboxylase antibodies (anti-GAD65) and islet cell antibodies (ICA) indicates a lack of insulin and/or relative need for insulin. Studies show:

  • In newly diagnosed patients with GAD65 antibodies, insulin dependency develops within 10 years in 50% of cases, whereas in patients without GAD65 antibodies it only develops in 3% of cases
  • In the UK Prospective Diabetes Study, 52% of GADA-positive patients became insulin dependent within 6 years
  • In a Finnish study, T2D patients with GAD65 antibodies had higher C-peptide levels and more symptoms of the metabolic syndrome than T1D patients with GAD65 antibodies. Genetic defects in insulin action

Rare genetic disorders that are accompanied by altered peripheral insulin action are usually associated with mutations of the insulin receptor /3/. Associated metabolic abnormalities can range from hyperinsulinemia and mild hyperglycemia to severe diabetes. Some individuals with these mutations may have hyperglycemia together with acanthosis nigricans, virilism or polycystic ovaries, or they may present with leprechaunism or Rabson-Mendenhall syndrome. Diseases of the exocrine pancreas

Any process that injures the pancreas and leads to a reduction in β-cell mass can cause diabetes. This applies to acquired processes such as pancreatitis, trauma, infection, pancreatectomy and pancreatic carcinoma. In pancreatic cancer (adenocarcinoma) other processes appear to play a role, because processes that involve only a small part of the organ are often already associated with diabetes /1/. The prevalence of diabetes in chronic pancreatitis is 30–70% /29/. It depends on the time of investigation, as shown in a study /30/ which reports an increase in incidence from 8% to 78% during a 10-year period of disease. The loss of exocrine and endocrine function does not occur at the same time. Patients with severe pancreatitis requiring enzyme substitution may have normal glucose tolerance. Hereditary hemochromatosis

The prevalence of diabetes is 50–60%. The severity depends on the degree of injury caused to the liver and pancreas. The decreased extraction of insulin from portal blood with consecutively reduced insulin clearance leads to the development of chronic hyperinsulinism. This causes a reduction of the tissue receptor sensitivity, leading to insulin resistance. As iron deposition in the Langerhans islets increases, β-cells are destroyed, resulting in insulin deficiency /31/. Endocrinopathies and diabetes

Hormones that antagonize insulin action can cause diabetes if they occur in elevated concentrations. This is the case, for example, with

  • Acromegaly (excess growth hormone)
  • Cushing’s syndrome (hypercortisolism)
  • Pheochromocytoma (catecholamine excess).

Hypercortisolism-induced and growth hormone-induced hypokalemia can cause diabetes by inhibiting insulin secretion. Hyperglycemia generally resolves after successful treatment of the underlying disease. Drug- or chemical-induced diabetes

Drugs and chemicals that have a toxic effect can cause direct injury to the β-cells and permanently destroy them. This is the case with the rat poison vacor or intravenous administration of pentamidine. More commonly, however, medications cause diabetes in individuals with existing insulin resistance by impairing insulin action. Examples are glucocorticoids, nicotinic acid, and interferon-α /2/. Infection-induced diabetes

Generalized viral infections can sometimes cause sufficient inflammatory injury to the pancreas to lead to the development of diabetes. The infections can be caused by Coxsackievirus B, Adenovirus, Cytomegalovirus, and the Mumps virus. Congenital Rubella virus infection, in contrast, causes type 1A diabetes /3/. Other genetic syndromes sometimes associated with diabetes

Many genetic syndromes can be associated with diabetes mellitus. Examples are listed in Tab. 3.1-1 – Classification of diabetes mellitus /3/.

Comments, criteria and laboratory findings for these types of diabetes are summarized in Tab. 3.1-5 – Laboratory findings in other types of diabetes.

3.1.6 Hyperglycemia during pregnancy

In the case of hyperglycemia during pregnancy, a distinction is made between:

  • Gestational diabetes
  • Women with undiagnosed diabetes type 2 who become pregnant. Gestational diabetes mellitus (GDM)

GDM is initially diagnosed during pregnancy at 24–28 weeks of gestation and is clinically not overt diabetes mellitus. The condition is associated with complications for the mother, the fetus and the newborn. The medical decision-making approach is as follows /38/:

  • During the first prenatal visit, a screening test for diabetes mellitus is performed. Pregnant women with the criteria (Tab. 3.1-6 – Testing for diabetes in pregnant women) have overt diabetes and will receive treatment /3/.
  • If the test results are not diagnostic of diabetes, a 75-g oGTT is performed at 24–28 weeks of gestation and evaluated according to the criteria in Tab. 3.1-6. A pathologic oGTT indicates GDM and the pregnant women will receive treatment.

Women with a history of GDM have a greatly increased subsequent risk for diabetes and should be followed up, to be reclassified into one of the following categories /38/:

  • Normoglycemia
  • Elevated fasting glucose
  • Impaired glucose tolerance
  • Diabetes mellitus.

According to studies in the U.S., diabetes complicates an average 7% of pregnancies, with prevalence varying between 1% and 14% depending on the population studied and the diabetes criteria selected. The risk is higher in pregnant women over 25 years and even higher in those with a positive family history. 88% of pregnant women with hyperglycemia have GDM, the rest have overt diabetes, thereof 35% type 1 and 65% type 2 /39/. GDM patients with risk factors (positive family history, T2D, body mass index above 27 kg/m2, over 35 years of age, GDM, preeclampsia, malformation, macrosomia or intrauterine fetal death in the previous pregnancy, hypertension or macrosomia in the current pregnancy) are at significantly higher risk of intrauterine fetal death, premature delivery, or elective cesarian section. Pathophysiology of GDM

Pregnancy has a profound effect on carbohydrate metabolism. The developing fetus relies on maternal supply of glucose, amino acids and lipids, which is primarily regulated by insulin. In the first trimester, during which the levels of human chorionic gonadotropin (hCG), estrogens and progesterone are increased, insulin sensitivity of the tissues is normal or increased. During the course of the pregnancy the woman becomes more insulin resistant, resulting in increased insulin secretion /40/. Insulin resistance is highest in the third trimester and is due to the increase in progesterone, prolactin, cortisol and human placental lactogen (hPL) levels. The main cause of insulin resistance is likely to be hPL as its structure is nearly homologous to that of growth hormone, which acts as an antagonist to insulin /41/. To maintain glucose homeostasis, more insulin must be secreted in order to overcome insulin resistance. Insulin resistance is linked to an increase in free fatty acids which intensify insulin resistance. The main cause that triggers GDM in some pregnant women appears to be a deficient reserve of insulin. Pregnant women with GDM exhibit a diminished first-phase insulin response to glucose load /40/. The similarities between GDM and T2D suggest that GDM is a prodromal form of type 2 diabetes being unmasked by pregnancy. Maternal risks and complications

Women with GDM are at increased risk of miscarriage due to hyperglycemia. After the pregnancy they are at increased risk of developing diabetes, in particular type 2. Overweight and other risk factors for insulin resistance increase the likelihood of developing T2D while the presence of islet cell antibodies increases the probability of developing T1D and LADA /42/. GDM patients should have their carbohydrate metabolism evaluated with the 75-g oGTT 6–12 weeks following delivery and then as described in Tab. 3.1-2 – Diagnosis of prediabetes and diabetes mellitus based on ADA criteria for diabetes screening.

Most women with GDM return to normoglycemia within a short period after childbirth. However, some will develop GDM again in a subsequent pregnancy.

Approximately 20% of patients with GDM still have impaired glucose tolerance during the early postpartum phase, and 17–63% will develop manifest diabetes within the following 5–16 years. In a Swedish study /43/, 30% of patients with GDM had developed diabetes and 51% had developed impaired glucose tolerance within 5 years post partum. Independent predictors in pregnancy were fasting blood glucose levels > 94 mg/dL (5.2 mmol/L) and HbA1c levels ≥ 5,7%.

Risk factors for the subsequent development of impaired glucose tolerance in pregnant women include /40/:

  • Insulin requirement during pregnancy
  • Elevated fasting glucose levels during pregnancy and post partum
  • Diagnosis of GDM during early pregnancy
  • Maternal overweight. Pregnancy with overt diabetes

Women with T1D or T2D who are planning to become pregnant have to consider the effect of the pregnancy on their diabetes and vice versa /41/.

Maternal risks and complications

Since about two-thirds of pregnancies in women with diabetes are unplanned, diabetics of childbearing age should be counseled at an early stage to plan for a desired pregnancy. To ensure a complication-free pregnancy and optimal development of the embryo and fetus, good glycemic control in the preconceptional phase and during pregnancy is of paramount importance.

During the 1st trimester of pregnancy in insulin-dependent diabetic women, the need for insulin usually decreases and there is a tendency to develop hyperglycemia. From the 2nd trimester, however, insulin demand increases steadily up to the preconceptional level. Laboratory diagnostics

To evaluate the metabolic situation and possible diabetes-related complications of the pregnant women, the following laboratory tests should be performed /42/:

  • Blood glucose measurements. The target levels are listed in Tab. 3.1-6 – Testing for diabetes in pregnant women. T2D patients receiving oral antidiabetic treatment and dietary counseling should be changed to insulin therapy, if they do not reach these targets. Blood sugar should be tested by a doctor to validate the patient’s self-monitoring results. The frequency of hypo- or hyperglycemic episodes must be inquired from the patient and investigated. Severe, frequent or inexplicable hypoglycemia can be caused by deficient counter regulation, impaired awareness, insulin medication errors, and excessive alcohol consumption.
  • Ketone bodies in urine and/or blood. The elevated concentration of ketone bodies in diabetic women with hyperglycemia is indicative of the onset or presence of ketoacidosis. Ketone bodies are indicative of absolute or relative insulin deficiency. Testing for ketone bodies is of particular importance in diabetic women who self-monitor their carbohydrate metabolism and in those with blood glucose levels > 200 mg/dL (11.1 mmol/L). Pregnant women with higher levels than these may have diabetic ketoacidosis, which is linked to a high fetal mortality rate.
  • HbA1c test. For preconceptional metabolic control, this test should be performed every four weeks, and then every 6–8 weeks once glycemic control has been achieved. Good control means levels below 6%; the target level is below 7%. During pregnancy, the HbA1c level should be determined monthly and should be below 6% /3/.
  • Serum creatinine to evaluate glomerular function. Due to the elevated filtration rate during pregnancy, the kidneys are more susceptible to dysfunction. Pregnancy is not advised in women who have diabetic nephropathy with a creatinine clearance < 50 [mL × min.–1 × (1.73 m2)–1] or serum creatinine > 3.0 mg/dL (51 mmol/L) prior to conception /38/.
  • Urinary albumin in relation to creatinine excretion to evaluate the effect of pregnancy on renal function. If renal function is normal prior to conception, there will be no renal dysfunction during pregnancy, with the exception of mild proteinuria, which will normalize post partum. If persistent albuminuria (30–300 mg/24 h; 20–200 μg/min.; 30–300 mg/g creatinine) is present prior to conception, about one-third of women with this condition will progress to proteinuria with excretions in the range of g/L by the end of the pregnancy. If persistent albuminuria (> 300 mg/24 h, > 200 μg/min., > 300 mg/g creatinine) is present prior to conception, the likelihood of preeclampsia developing close to the date of childbirth is 30% /46/. Some women develop retinopathy and nephrotic proteinuria /44/.
  • Investigation for hypertension, which is a common concomitant complication in diabetic pregnancies. This applies in particular if persistent albuminuria is present prior to conception. Pregnant women with T1D usually develop hypertension in combination with diabetic nephropathy, which is accompanied by persistent albuminuria. Pregnant women with T2D are more likely to have hypertension than those with T1D /44/.
  • TSH in T1D; 5–10% of these women have hypo- or hyperthyroidism /47/. Fetal risks and neonatal complications in diabetic pregnancies

Diabetic embryopathy

During organogenesis, a diabetic metabolism increases the rate of miscarriage. The risk increases linearly with the hyperglycemia measurable based on the HbA1c level, and the number of damaged organs increases with increasing maternal glucose levels. Anomalies in diabetic embryopathy include defects of the neural tube, omphalocele, musculoskeletal anomalies, deformities of the kidneys and urinary system, and conotruncal heart defects. The increased rate of malformations contributes significantly to perinatal mortality [Death between pregnancy week 20 (22) and 7th day of life].

Since organ development is not complete until about 6 weeks post conception, good diabetic control prior to conception is important. Ideally, the HbA1c level prior to conception should be < 7%, or better < 6% /3/. Since gestational diabetes does not develop until the second half of pregnancy, impaired glucose tolerance in the earlier months is uncommon and therefore does not lead to congenital malformations. Where such malformations are reported, T2D is usually present. T2D is becoming more prevalent in younger pregnant women. Although glucose levels in these women are more stable than in those with T1D, hyperglycemia is nevertheless present and there is a risk of congenital malformations.

Diabetic fetopathy

Maternal hyperglycemia in the second half of pregnancy causes diabetic fetopathy /41/. In addition to macrosomia, respiratory distress syndrome and postnatal hypoglycemia, main findings also include hyperbilirubinemia, polyglobulia, hypocalcemia, and hypogmagnesemia (Tab. 3.1-7 – Tests for the risk assessment of infants born to diabetic mothers).

Macrosomia: this abnormal condition is defined as a birth weight > 90th percentile of the gestational age. Since insulin stimulates tissue growth, the increased fetal insulin production associated with the mother’s diabetes leads to increased growth, in particular of the trunk. The main determinants of macrosomia are maternal weight and maternal weight gain, gestational age, and maternal glucose concentrations, the latter being the main treatable determinant of macrosomia. Approximately 20% of untreated diabetic women give birth to a macrosomic child. For fetuses of pregnant women with T1D the risk of being macrosomic is 25% and that of having hyperglycemia is 8% /41/.

Neonatal hypoglycemia: this state results from hyperplasia of the islet organs due to increased glucose supply by the diabetic mother. Even though newborns can have glucose levels < 45 mg/dL (2.5 mmol/L), 5–24% of infants born to diabetic mothers still have lower levels during the neonatal period (see also Tab. 3.2-2 – Hypoglycemia syndromes in childhood and infancy). Neonatal hypoglycemia can cause subsequent disorders of the central nervous system.

Respiratory distress syndrome: the syndrome is due to delayed pulmonary maturity and has an incidence of 1.6%, which is comparable to that of nondiabetic pregnancies. Neonatal diabetes mellitus (NDM)

Uncontrolled hyperglycemia within the first 6 months of life occurs in all races and ethnic groups /52/. The majority of newborns present with intrauterine growth retardation, failure to thrive, lack of subcutaneous fat, and low or undetectable C-peptide levels. In most cases of NDM, the disease is monogenic and there is no evidence of islet cell autoantibodies. There are mutations in genes that are responsible for the development of the pancreas, β-cell apoptosis and the regulation of insulin production. The incidence is about 1 per every 300,000 to 500,000 live births.

A differentiation must be made between the above described permanent neonatal diabetes mellitus (PNDM) and the transient form (TNDM). In a large proportion of infants with TNDM, blood glucose returns to normal within the first few months, but they go on to develop T2D years after this initial hyperglycemia. About 57% of NDM cases belong to the TNDM group. They initially require treatment with insulin, but this can usually be discontinued within less than 18 months.

3.1.7 Polycystic ovary syndrome

Polycystic ovary syndrome (PCOS) is a heterogeneous disorder which affects 5–10% of women of childbearing age /53/. It manifests clinically as chronic anovulation with oligo-/amenorrhea, infertility, and hyperandrogenism. 40–50% of women with PCOS are insulin resistant and may develop symptoms of metabolic syndrome along with cardiovascular disease, hypertension, vascular dysfunction and obstructive sleep apnea in addition to having an increased prevalence of endometrial carcinoma.

PCOS is a prediabetic state with a 31–35% prevalence of impaired glucose tolerance (IGT) and a 7.5–10% prevalence of T2D /53/. The rate of conversion from IGT to T2D is reported to be 5–10 times higher in women with PCOS than in those without this syndrome /54/. Insulin resistance is assessed using the Homeostasis Model Assessment (HOMA) test (see Tab. 3.7-6 – Interpretation of baseline and glucagon-stimulated C-peptide levels in diabetics).

3.1.8 Tests for glycemia in diabetes

The determination of blood glucose levels is the most reliable test for hyperglycemia.

Testing for diabetes in asymptomatic patients

The following tests are indicated when diabetes is suspected (Tab. 3.1-8 – Testing for diabetes and glucose monitoring):

  • Fasting plasma glucose (FPG)
  • Random plasma glucose
  • 2-hour plasma glucose during 75-gram oral glucose tolerance test
  • HbA1c in blood
  • Ketone bodies in blood or urine.

Classification of diabetes mellitus

The following tests are of importance:

  • Antibodies against epitopes of the β-cells such as anti-insulin, anti-GAD 65 and anti-tyrosine phosphatase IA-2 and IA-2 b
  • Genetic markers (risk alleles, HLA markers)
  • C-peptide, insulin, proinsulin in plasma.

Once diabetes has been diagnosed, the determination of insulin, proinsulin and C-peptide plays a minor role in routine diagnosis.

Diagnosis of prediabetes

The following tests are indicated if prediabetes is suspected:

3.1.9 Glycemic control in diabetics

The diagnostic symptoms of diabetes are persistent hyperglycemia and an elevated HbA1c. Blood glucose must be lowered to near-normal levels to prevent the following complications /1/:

  • Impaired vision, polyuria, polydipsia, fatigue, weight loss with polyphagia, vaginitis, or balanitis
  • Risk of hypoglycemia, acute hyperglycemia in the form of diabetic ketoacidosis, hyperglycemic hyper osmolar non ketotic syndrome. Hyperglycemia is associated with increased morbidity and mortality.
  • Risk of cardiovascular disease, diabetic retinopathy, nephropathy, and neuropathy.
  • A lipid profile that is not associated with increased atherogenic risk.

The targets and concepts for the control of glycemia, lipidemia and blood pressure are shown in Tab. 3.1-9 – Glycemic goals in the treatment of diabetes. Management of glycemic control

Self-monitoring of blood glucose and measurement of HbA1c are important components in the glycemic control of diabetics.

Self-monitoring of glucose (SMBG): SMBG allows patients to monitor their individual response to treatment, to check whether glycemic goals are reached, and to prevent hypoglycemia under therapy. Recommendations for SMBG are as follows /1/:

  • TD1, TD2 and GDM on multiple-dose insulin: 3–4 times daily prior to meals
  • Patients who inject insulin only occasionally, take oral anti diabetics or control their diabetes through diet: occasionally for monitoring, mainly after meals.

Continuous glucose monitoring (CGM): an international panel of physicians, researchers, and individuals with diabetes recommended CGM as a robust research tool, and continuous glucose data should be recognized by governing bodies as a valuable and meaningful end point to be used in clinical trials of new drugs and devices for diabetes treatment /86/.

Determination of HbA1c: the purpose of the measurement is to estimate the mean glucose level of the past 2–3 months. The HbA1c measurement at the beginning of treatment serves as the baseline value of glycemia. Testing at least twice per year is a criterion of whether metabolic control of diabetes has been achieved and whether the level is within the target range. The individual target of a diabetic should be as close as possible to the upper reference limit of 6% without provoking frequent hyperglycemic episodes. The relationship between plasma glucose levels averaged over a 2–3 month period and HbA1c is shown in Tab. 3.1-10 – Correlation of plasma glucose with the HbA1c level.

HbA1c monitoring /2/:

  • At least two times a year in patients who are meeting treatment goals (and who have stable glycemic control)
  • Quarterly in patients whose therapy has changed or who are not meeting glycemic goals. Benefits of good glycemic control

The Diabetes Control and Complication Trial (DCCT) demonstrated that a regimen of intensive therapy aimed at maintaining near-normal glycemic control reduces the risk of development or progression of retinopathy, nephropathy and neuropathy in T1D by 50–70% /1/. This regimen achieved a median HbA1c of 7.2% compared with conventional therapy with a median HbA1c of 9%. The DCCT reference interval for HbA1c was 4–6%.

For T2D, the United Kingdom Prospective Diabetes Study (UKPDS) demonstrated that improved glycemic control reduces the risk of developing retinopathy, nephropathy and possibly neuropathy. With tighter glycemic control, the overall microvascular complication rate was decreased by 25% compared with conventional treatment. Epidemiological analysis showed a continuous relationship between hyperglycemia and the rate of microvascular complications, such that for every percentage point decrease in HbA1c (e.g., 9% to 8%), there was a 35% reduction in the risk of microvascular complications.

3.1.10 Acute complications of diabetes

Diabetes is associated with acute and chronic complications. In addition, metabolic stress responses can be expected, which occur with acute diseases.

The main acute complications of diabetes are hypoglycemia, metabolic stress response, diabetic ketoacidosis, and hyperglycemic hyper osmolar non ketotic syndrome.

3.1.11 Iatrogenic hypoglycemia

In healthy individuals, regulating mechanisms that increase glucose levels are activated when blood glucose falls below 70 mg/dL (3.9 mmol/L). They prevent hypoglycemia and insufficient supply of the brain with high-energy substrate. In people with diabetes therapeutic insulin excess caused by treatment with insulin, sulfonylureas and glinides can initiate hypoglycemic episodes /58/. Defective glucose counter regulation

Typically hypoglycemia occurs during less marked or even relative therapeutic insulin excess in patients with diminished exogenous glucose supply, decreased endogenous glucose production, increased glucose consumption or increased insulin sensitivity. Such patients, those with T1D or advanced T2D have β-cell failure or absolute insulin deficiency and compromised defenses against hypoglycemia. As glucose levels fall, the compromised physiologic defenses include failure of insulin levels to fall, failure of glucagon secretion to increase and attenuated epinephrine secretion. This combination of compromised physiologic defenses cause the syndrome of defective glucose counter regulation with increased risk of recurrent hypoglycemia. An important finding is that prior hypoglycemia weakens the body’s defense against subsequent hypoglycemia in type 1 diabetics and non diabetics. This led to the concept of hypoglycemia associated autonomic failure (HAAF). According to this concept recent hypoglycemia causes both defective counter regulation and hypoglycemia unawareness. Rate of hypoglycemia

The rate of hypoglycemia is about 10 times higher in T1D than in T2D. The rate of severe hypoglycemia (requiring the assistance of another person) is 62–170 episodes per 100 years in T1D and 3–73 in T2D. While in T1D glucose counter regulation is impaired at an early stage, in T2D it will become compromised only in absolute insulin deficiency. The risk of iatrogenic hypoglycemia is similar to that in T1D. Children under 6–7 years of age usually have hypoglycemia unawareness. The ADA recommends treatment with glucose or carbohydrate-containing foods if glucose levels are below 70 mg/dL (3.9 mmol/L).

Nocturnal hypoglycemia is a significant problem. In the Diabetes Control and Complication Trial, more than half of severe hypoglycemic events in T1D occurred during sleep. They were caused by the glucose-lowering effects of evening exercise, sleep-induced defects in counter regulatory hormone responses to hypoglycemia, and missed bedtime snacks. A 12-month study /59/ of children and adults with T1D showed the following findings:

  • 7.4% of participants had hypoglycemic events during 8.5% of nights
  • The duration of hypoglycemia was longer than 2 h in 23% of nights with hypoglycemia
  • Hypoglycemia was defined as two consecutive glucose readings ≤ 60 mg/dL (3.3 mmol/L) in 20 minutes. Metabolic stress response

Surgery or other stressful events can induce a metabolic stress response in critically ill patients, which is caused by the increase of insulin counter regulatory hormones such as noradrenaline, glucagon, cortisol and growth hormone. These hormones have a catabolic effect and increase gluconeogenesis and lipolysis, leading to elevated levels of glucose, free fatty acids and ketone bodies in blood. The increase in these substrates impairs the insulin secretory response of the islet cells. Combined with acidosis, which may develop due to the increased levels of lactate and ketone bodies, this will result in lower insulin sensitivity and increasing insulin resistance of the tissues, thereby worsening the metabolic situation.

To differentiate diabetes from stress-induced hyperglycemia, glucose and HbA1c concentrations should be determined. An elevated HbA1c level in the presence of hyperglycemia is indicative of pre-existing diabetes while a normal level generally precludes it.

More than 90% of critically ill patients have hyperglycemia (glucose > 126 mg/dL; 7.0 mmol/L) due to a stress response elicited by surgery or other events.

For adults, the NICE-Sugar Study /60/ demonstrated that tight insulin-based glycemic control with a target blood glucose range of 81–108 mg/dL (4.5–6.0 mmol/ L) is associated with higher mortality than glycemic control with a target of ≤ 180 mg/dL (10.0 mmol/L).

For children aged 0–3 years who have undergone heart surgery, insulin-based glycemic control with a target blood glucose range of 81–108 mg/dL (4.5–6.0 mmol/L) showed no benefit compared to higher levels. The length of hospitalization, the rate of infections and mortality are the same, but the rate of hypoglycemic events was significantly higher.

3.1.12 Diabetic ketoacidosis (DKA)

DKA is defined as the occurrence of metabolic acidosis with ketonemia. The main symptoms are ketonemia/ketonuria, metabolic acidosis and dehydration. In 35–40% of children, DKA is detected at diagnosis of T1D. While blood glucose concentrations in individuals with DKA are usually in the range of 400–500 mg/dL (22.2–27.8 mmol/L), it has been shown that some individuals may have levels under 300 mg/dL (16.7 mmol/L), above 800 mg/dL (44.4 mmol/L), or even within the reference interval /61/. The latter may be the case when there is severe dehydration with a reduced glomerular filtration rate. Hyperglycemia causes dilutional hyponatremia. For every 100 mg/dL (5.6 mmol/L) increase in glucose, the sodium concentration decreases by 1.6 mmol/L /61/. Further information on DKA can be found in Section 5.5 – Ketone bodies.

3.1.13 Hyperglycemic hyperosmolar non ketotic syndrome (HHNS)

HHNS is an acute, life-threatening complication of diabetes. It is caused by a relative or absolute insulin deficiency and elevated levels of insulin counter regulatory hormones such as glucagon, catecholamines, growth hormone, and cortisol. Although HHNS may occur in both T1D and T2D, DKA is generally associated with T1D, and HHNS with T2D /62/. Patients are often unaware of having diabetes. HHNS mainly affects individuals over the age of 55. Blood glucose is usually above 600 mg/dL (33.3 mmol/L) and serum osmolality is elevated above 330 mmoL/kg. In recent years, HHNS has been increasingly diagnosed in children with T2D. In these cases, T2D initially manifests with symptoms and findings of HHNS. It mainly affects overweight children aged 10 years and older of African-American descent /63/. Further information on HHNS can be found in Section 5.5 – Ketone bodies.

3.1.14 Chronic complications of diabetes

Pathophysiological mechanisms associated with diabetes comprise endothelial vascular dysfunction, low-grade inflammation, and thrombocyte dysfunction.

There are three main categories of chronic diabetes complications:

  • Microvascular complications, in particular retinopathy and nephropathy.
  • Macro vascular complications, in cardiovascular disease, cerebrovascular disease, and peripheral vascular diseases.
  • Neuropathies of both the peripheral and the autonomic nervous system. Microvascular disease

Even though diabetics usually die from macro vascular complications, microvascular complications such as retinopathy and nephropathy play an important role because they significantly restrict the quality of life. Microvascular disease develop as a result of poor metabolic control.

Early changes include hyperglycemia-induced vascular dilatation with increased blood flow, as well as increased intravascular pressure in the capillaries of the retina and the renal glomeruli. As a result, there is increased leakage of proteins from the capillaries, measurable by the presence of micro albuminuria.

Long-term hyperglycemia-induced changes include altered structuring of the extracellular matrix, in particular the basal membrane of the vessels. The toxic effect of glucose is caused by various mechanisms /59/:

  • The direct i.e., not enzyme-mediated, reaction of glucose with proteins. This reaction, also known as non-enzymatic glycosylation, leads to increased glycation, in particular of long-life proteins such as collagen.
  • The glycation of proteins is the starting point for the gradual formation of advanced glycation end products (AGEs), which are responsible for the vascular damage that is caused when the AGEs bind to the AGE receptors in the vascular cells, activating the synthesis of inflammatory cytokines (also refer to Section 3.6 – Hemoglobin A1c).

Glycemic control /1/: strict glycemic control plays an important role in retarding the development and progression of microvascular complications. The Epidemiology of Diabetes Interventions and Complications (EDIC) study showed that tight glycemic control significantly reduced the progression of retinopathy and nephropathy in patients with T1D. This was also the case in the UK Prospective Diabetes Study (UKPDS) of patients with T2D and in the Veterans Affairs Diabetes Trial (VADT), in which glycemic control with a mean HbA1c of 6.9% resulted in a significant reduction in micro albuminuria. The Action in Diabetes and Vascular Disease (ADVANCE) study similarly showed a significant reduction in albuminuria in patients with T2D when the mean HbA1c was reduced from the general goal of 7.0% to the stringent goal of 6.3%. Diabetic retinopathy

Diabetic retinopathy is a specific microvascular complication of diabetes. In the industrialized countries it is the leading cause of blindness among adults aged 20–65 /64/. The risk of developing diabetic retinopathy after 20 years of diabetes is nearly 100% in T1D and over 60% in T2D. Diabetic retinopathy progresses /65/:

  • From a mild non-proliferative form, characterized by increased vascular permeability
  • To moderate and severe non-proliferative retinopathy, characterized by vascular closure
  • To proliferative retinopathy, characterized by the growth of new blood vessels on the retina and posterior surface of the vitreous.

A significant factor contributing to the pathophysiology of the disease is the activation of angiotensin II in the retina where it mediates vascular growth and accelerates the development of proliferative retinopathy. The permeability of the retinal capillaries for proteins is increased, thereby promoting the development of macular edema.

The DCCT showed that stringent glycemic control can reduce or prevent the development of diabetic retinopathy by 27% and the progression of it in 34–76% of cases. This improvement was achieved with an average 10% reduction in HbA1c from 8% to 7.2% (upper reference limit 6%). The UKPDS showed that for every percentage point decrease in HbA1c there was a 35% reduction in the risk of microvascular complications /65/. Diabetic nephropathy

20–30% of patients with T1D develop diabetic nephropathy about 20 years following diagnosis of the disease. In T2D, the prevalence is lower, but due to the higher prevalence of T2D, 60% of diabetics with end stage renal disease (ESRD) belong to this type. African Americans and Hispanics with T2D progress to dialysis earlier than non-Hispanic Caucasians. The number of diabetics requiring dialysis is on the rise, since the prevalence of T2D is increasing and these patients increasingly live longer. However, at the stage of chronic renal failure, only 20% of patients have a life expectancy of more than 5 years /66/.

In addition to hyperglycemia, the increased production of angiotensin II also plays an important role in diabetic microangiopathy. It causes contraction of the efferent arterioles in the kidney, which increases the filtration pressure in the glomeruli, resulting in increased excretion of albumin. Angiotensin II also increases systemic blood pressure and induces endothelial dysfunction and glomerular injury.

The earliest evidence of nephropathy is albuminuria. Normoalbuminuria is defined as an excretion rate of < 30 mg/24 h.

Diabetic nephropathy develops in the following stages /66/:

  • Persistent albuminuria with an excretion rate of 30–299 mg/24 h. This is the stage of incipient diabetic nephropathy. It is usually accompanied by glomerular hyper filtration and early hypertension. Albuminuria is classified as persistent if it is detectable in at least two out of three urine samples within 6 months.
  • Persistent albuminuria with an excretion rate of ≥ 300 mg/24 h. This is the stage of early overt diabetic nephropathy. Albuminuria ≥ 300 mg/24 h is also referred to as clinical albuminuria. Hypertension is also present. In T1D this stage develops untreated over a period of 10–15 years following diagnosis of persistent albuminuria, with albuminuria increasing at a rate of 10–20% per year. Only 20–40% of T2D patients progress to this stage.
  • Advanced diabetic nephropathy. This stage is characterized by the progressive increase of proteinuria and hypertension. The glomerular filtration rate (GFR) gradually falls over a period of several years at a rate that shows yearly inter individual variability of 2–20 [mL × min.–1 × (1.73 m2)–1]. The decreasing GFR can be detected early by determining the creatinine or cystatin C based estimated GFR.
  • End stage renal disease (ESRD). Without treatment, 50% of type 1 diabetics with clinical nephropathy will progress to ESRD within 10 years, and over 75% within 20 years, whereas in T2D, only 20% of diabetics with clinical nephropathy will develop ESRD within 20 years.

It must be noted that transient albuminuria can also be caused by exercise, urinary tract infections, short-term hyperglycemia, marked hypertension, heart failure, and acute febrile illness.

Persistent albuminuria is not only the earliest indicator of diabetic nephropathy, but also a marker of elevated cardiovascular morbidity and mortality of diabetics.

Early therapeutic intervention in diabetics can delay the development of renal complications and reduce the progression of the disease. The UKPDS, ADVANCE, and STENO-2 studies have shown that stringent blood glucose and blood pressure control reduce the incidence and progression of diabetic nephropathy. In patients with T2D, inhibition of the renin-angiotensin-aldosterone system by angiotensin-converting enzyme (ACE) inhibitors or an ACE receptor blockers delayed the occurrence of albuminuria and retarded the progression to ESRD. The use of ACE inhibitors and ACE receptor blockers is therefore part of the standard treatment of patients with T2D. Also refer to Section 12 – Kidney and urinary tract.

End stage renal disease

The overwhelming body of evidence on glycemic control in end stage renal disease has been obtained using HbA1c /85/. Macrovascular disease

Macro vascular diseases in patients with T1D and T2D include cardiovascular disease, cerebrovascular and peripheral vascular diseases. If diabetes is diagnosed, life expectancy is reduced by 30%, with cardiovascular disease being the leading cause of death. In Sweden from 1998 through 2014, mortality and the incidence of cardiovascular outcomes declined. Patients with T1D had roughly 40% greater reduction in cardiovascular outcomes than controls and patients with T2D had roughly 20% greater reductions than controls. Reductions in fatal outcomes were similar in patients with diabetes and controls /67/. Cardiovascular disease (CVD)

The reduction in life expectancy in diabetics is mainly due to cardiovascular events. The risk for CVD is 2–3 times higher in male diabetics and even 3–5 times higher in female diabetics than in non diabetics of the same age. An increased risk for CVD exists even before T2D is diagnosed. 40% of patients with newly diagnosed T2D have CVD, and 80% of patients with CVD have T2D or prediabetes. More than 60% of patients with T2D will die of CVD /168/. Interventions such as change of lifestyle, control of blood pressure, reduction of lipids, and treatment with anti-platelet agents can reduce the progression and complications of T2D. Results of studies investigating the effect of strict glycemic control have been unsatisfactory, but data from the UKPDS have shown that it has a protective effect /6970/. In the ADVANCE, VADT and Action to Control Cardiovascular Disease in Diabetes (ACCORD) studies, however, strict glycemic control was associated with increased mortality and an increased risk for CVD.

Myocardial infarction (MI)

Diabetics with AMI and a glucose levels above the range of 124–180 mg/dL (6.9–10.0 mmol/L) at admission are at nearly twice the risk for congestive heart failure, cardiogenic shock or in hospital death compared to non diabetics /71/. If a patient with AMI has a blood glucose level > 180 mg/dL (10.0 mmol/L) at admission, this is usually not due to stress-induced hyperglycemia, but undiagnosed diabetes /60/. It is important that hyperglycemia in patients admitted with AMI is normalized by appropriate treatment. The Diabetes Insulin Glucose in Acute Myocardial Infarction (DIGAMI) study has shown that immediate reduction of blood glucose levels to below 198 mg/dL (11.0 mmol/L) by insulin treatment reduces mortality by about 30% /72/.

Coronary bypass surgery

According to data from the National Cardiac Surgery Database (NCSD), diabetics undergoing bypass surgery have an approx. 50% higher mortality rate than non diabetics /60/. Neuropathies and cerebrovascular disease

Neurological complications of diabetes include neuropathy, stroke and Alzheimer’s disease.


The diabetic neuropathies are heterogeneous with diverse clinical manifestations. There are two main types of neuropathy /3/:

  • Distal symmetric polyneuropathy (DPN). There is progressive loss of myelinated nerve fibers, and segmental demyelination develops. Early detection is important since stringent diabetic control retards the progression. Moreover, DPN is asymptomatic in up to 50% of cases. There may be painless numbness starting from the toes and feet, and impaired perception of pain and temperature with the risk of painless ulcers developing. Data from the DCCT show that strict diabetic control can reduce the development and progression of clinical neuropathies by 64% in T1D and by 42% in T2D /73/.
  • Autonomic neuropathy. It can affect any organ and system in the body. Clinical symptoms include tachycardia at rest, exercise intolerance, orthostatic hypotension, constipation, gastroparesis, erectile dysfunction, and urination problems. Autonomic neuropathy is also a risk factor for CVD.


According to epidemiological studies, the prevalence of thromboembolic, not hemorrhagic, stroke is 2–6 times higher in diabetics than in non diabetics. In the Framingham study, the incidence of stroke was 3.6 times higher in diabetic women and 2.5 times higher in diabetic men than in non diabetics of the same age group /74/.

The following relationships exist between blood glucose levels and stroke /71/:

  • Hypoglycemia worsens the prognosis for stroke. In the presence of global ischemia, glucose concentrations below 65 mg/dL (3.6 mmol/L) lead to higher mortality and a worse functional outcome /75/.
  • Acute hyperglycemia increases mortality and worsens the prognosis for cerebral hemorrhages.

Alzheimer’s disease

Patients with T2D have a 2–2.5-fold higher risk of developing Alzheimer’s disease and vascular dementia than non diabetics of the same age group. The cause is reported to be microvascular infarctions in subcortical structures /76/. Arterial thrombosis

In diabetes there is an imbalance between pro coagulants and the factors of fibrinolysis, leading to the deposition of fibrin clots along the vessel wall. This is thought to induce the formation of arterial thrombi even before the endothelium is compromised and structural damage is caused to the intima. The formation of fibrin clots activates the release of mitogens and mediators of inflammation which then cause alterations in the intima /77/. Therefore, as an example, the prognosis for stroke patients with prediabetes and T2D is worse in comparison to patients with normoglycemia, and T2D is reported to double the risk of stroke /78/. The risk of mortality also is reported to be twice as high if venous plasma glucose levels are > 144 mg/dL (8.0 mmol/L) at admission, irrespective of age, type and severity of the stroke /78/. Infection

Diabetes is a risk factor for infections such as wound infections after surgery. The incidence of infections is 2–5 times higher in diabetics than in non diabetics. The rate of wound infections is higher (compared to a lower glucose level) if the mean venous plasma glucose level is above 200 mg/dL (11.1 mmol/L) within 36 h following surgery. Diabetics with glucose levels > 220 mg/dL (12.2 mmol/L) on the first day after surgery are 2.7 times more likely to have nosocomial infections with the clinical symptoms of sepsis, pneumonia, and wound infection than those with lower levels /79/. Cancer

Type 2 diabetes is associated with an increased risk for breast carcinoma, hepatocellular carcinoma and carcinomas of the gallbladder, colon, and endometrium /80/.

3.1.15 Drug-associated side effects in diabetes

The risk of hypoglycemia or hyperglycemia in diabetes increases exponentially with the number of drugs prescribed to the diabetic /81/. Important interactions of drugs that cause reduced glucose tolerance and hyperglycemia are listed in Tab. 3.1-11 – Drug- or chemical-induced diabetes; for drugs that induce hypoglycemia refer to Section 3.2 – Hypoglycemia syndromes.

3.1.16 Autoimmune polyglandular syndromes and diabetes

The autoimmune polyglandular syndromes (APS) are characterized by the immune-mediated destruction of endocrine tissues. A distinction is made between type 1 (APS-1) and type 2 APS (APS-2).

APS-1 /83/

Autoimmune poly endocrinopathy-candidiasis-ectodermal-dystrophy syndrome (APECED) is characterized by mucocutaneous candidiasis, autoimmune destruction, in particular of the endocrine glands, and ectodermal dystrophy. APECED is a rare, autosomal recessive disease, which occurs with equal frequency in both sexes. It is caused by mutations in the autoimmune regulator (AIRE) gene on chromosome 21q22.3, which is involved in the induction and maintenance of immunotolerance. The most common clinical manifestations are hypoparathyroidism, candidiasis, Addison’s disease, alopecia, and hypogonadism. The incidence of T1D is 5–10%. Early symptoms occur in the first decade of life. Most patients have 3–5 disease components.

APS-2 /83/

This syndrome is characterized by the coexistence of Addison’s disease, autoimmune thyroid disease, and T1D. The prevalence of T1D is 50–60%. Other less common diseases associated with APS-2 include pernicious anemia, vitiligo, celiac disease, alopecia, gonadal insufficiency, hypophysitis, and myasthenia gravis. The prevalence of APS-2 is 15–45 per 1 million inhabitants. Women are affected 1.6–3 times more than men. The autoimmunopathies often manifest beginning from the second to the third decade of life. Biomarkers for the diagnosis of diabetes include screening for hyperglycemia and, if necessary, autoantibodies against islet cell antigens. Thyroid disease associated with the APS-2 syndrome can manifest as hypo- or hyperthyroidism. APS-2 is a genetic disease. It is thought to be caused by multifactorial genetic factors. There is a strong association with HLA B8, DRB1*0301 (DR3) and DQA1*0501-DQB1*0201 (DQ2).


1. Shaw JE, Sicree RA, Zimmet PZ. Global estimates of the prevalence of diabetes for 2010 and 2030. Diabetes Res Clin Pract 2010; 87: 4–14.

2. Classification and diagnosis of diabetes: Standards in medical care in diabetes – 2018.Diabetes Care 2018; 41, Suppl 1: S13-S27.

3. Standards of medical care in diabetes – 2014. Diabetes Care 2014; 37, Suppl 1: S14–S80 and S81-S90.

4. Mayer-Davis E, Lawrence JM, Dabelea D, Divers J, Isom S, Dolan L, et al. Incidence trends of type 1 and type 2 diabetes among youths, 2002–2012 N Engl J Med 2017; 376: 1419–29.

5. Sperling MA. Aspects of the etiology, prediction, and prevention of insulin-dependent diabetes mellitus in childhood. Pediat Clin North Am 1997; 44: 269–84.

6. Rathmann W, Scheidt-Nave C, Roden M, Herder C. Typ-2-Diabetes: Prävalenz angeborener and erworbener Faktoren für die Prädiktion. Dtsch Arztebl Int 2013; 110: 331–7.

7. Balasubramanyam A, Nalini R, Hampe CS, Maldonado M. Syndromes of ketosis-prone diabetes mellitus. Endocrine Reviews 2008; 29: 292–302.

8. Noble JA. Immunogenetics of type 1 diabetes: A comprehensive review. J Autoimmunity 2015; 64: 101–12.

9. Concannon P, Rich SS, Nepom GT. Genetics of type 1A diabetes. N Engl J Med 2009; 360: 1646–54.

10. She JX. Genetic factors in type 1 diabetes: a complex disease. J Clin Lig Assay 1998; 21: 272–81.

11. Karounos DG. Convergence of genetic and environmental factors in the immunopathogenesis of type 1 diabetes mellitus. J Clin Lig Assay 1998; 21: 261–71.

12. Buysschaert M, Bergman M. Definition of prediabetes. Med Clin N Am 2011; 957: 289–97.

13. Vidal J, Kahn SE. Regulation of insulin secretion in vivo. In Lowe WL, Jr, ed. Genetics of diabetes mellitus. Boston; Kluwer 2000: 109–31.

14. Kolb H, Mandrup-Poulsen T. The global diabetes epidemic as a consequence of lifestyle-induced low-grade inflammation. Diabetologia 2010; 53: 10–20.

15. Ward WK, Bolgiano DC, McKnight B, et al. Diminished B cell secretory capacity in patients with noninsulin-dependent diabetes mellitus. J Clin Invest 1984; 74: 1318–23.

16. Joost HG, Fritsche A, Häring HU, Pfeiffer AFH, Roden M, Schulze MB. Diabetes mellitus Typ 2: Risikobestimmung wird präzisiert. Dtsch Ärztebl 2010; 107: 511–3.

17. Kahn HS, Cheng YJ, Thompson TJ, Imperatore G, Gregg EW. Two risk-scoring systems for predicting incident diabetes mellitus in U.S. adults age 45 to 64 years. Ann Intern Med 2009; 150: 741–51.

18. Dinneen SH, Gerich J, Rizza R. Carbohydrate metabolism in non-insulin-dependent diabetes mellitus. N Engl J Med 1992; 327: 707–13.

19. Bartoli E, Castello L, Sainaghi PP, Schianca GPC. Progression from hidden to overt type 2 diabetes mellitus: significance of screening and importance of the laboratory. Clin Lab 2005; 51: 613–24.

20. Butler AE, Janson J, Bonner-Weir S, Ritzel R, Rizza RA, Butler PC. β-cell deficit and increased β-cell apoptosis in humans with diabetes type 2. Diabetes 2003; 52: 102–10.

21. Ahlqvist E, Ahluwaklia S, Groop L. Genetics of type 2 diabetes. Clin Chem 2011; 57: 241–54.

22. Eastman RC, Cowie CC, Harris MI. Undiagnosed diabetes or impaired glucose tolerance and cardiovascular risk. Diabetes Care 1997; 20: 127–8.

23. Holman RR. Assessing the potential for α-glucosidase inhibitors in prediabetic states. Diab Res Clin Pract 1998; 40 (Suppl): S21–S25.

24. Lamb EJ, Day AP. New diagnostic criteria for diabetes mellitus: are we any further forward? Ann Clin Biochem 2000; 37: 588–92.

25. Haataja L, Gurlo T, Huang CJ, Butler PC. Islet amyloid in type 2 diabetes, and their toxic oligomer hypothesis. Endocrine Reviews 2008; 29: 303–16.

26. Vaxillaire M, Froguel P. Monogenic diabetes in the young, phamacogenetics and relevance to multifactorial forms of the young. Endocrine Reviews 2008; 29: 254–64.

27. McDonald TJ, Ellard S. Maturity onset diabetes of the young: identification and diagnosis. Ann Clin Biochem 2013; 50: 403–15.

28. Grant SFA, Hakonarson H, Schwartz S. Can genetics of type 1 and type 2 diabetes shed light on the genetics of latent autoimmune diabetes? Endocrine Reviews 2010; 31: 183–93.

29. Raue G, Keim V. Sekundärer Diabetes bei chronischer Pankreatitis. Z Gastroenterol 1999; 37 (Suppl 1): 4–9.

30. Lankisch PG, Löhr-Happe A, Otto J, Creutzfeldt W. Natural course in chronic pancreatitis. Pain, exocrine and endocrine pancreatic insufficiency and prognosis of the disease. Digestion 1993; 54: 148–55.

31. Phelps G, Chapman I, Hall P, Braund W, Mackinnon M. Prevalence of genetic haemochromatosis among diabetic patients. Lancet ii 1989; 233–4.

32. Larsen S. Diabetes mellitus secondary to chronic pancreatitis. Danish Medical Bulletin 1993; 40: 153–62.

33. Müller MJ, Pirlich M, Balks HJ, Selberg O, Glucose intolerance in liver cirrhosis: role of hepatic and non-hepatic influences. J Clin Chem Clin Biochem 1994; 32. 749–58.

34. George DK, Evans RM, Crofton RW, Gunn RI. Testing for hemochromatosis in the diabetic clinic. Ann Clin Biochem 1995; 32: 521–6.

35. O’Halloran DJ, Shalet SM. Acromegaly: Unravelling a complex disease. Growth Regulation 1995; 5: 119–24.

36. Bowes SB, Benn JJ, Scobie IN, Umpleby AM, Lowy C, Sönksen PH. Glucose metabolism in patients with Cushing’s syndrome. Clin Endocrinol 1991; 34: 311–6.

37. Schlaghecke R. Diabetes mellitus bei verschiedenen endokrinologischen Erkrankungen. In: Berger M, ed. Diabetes mellitus. München; Urban and Schwarzenberg 1995: 458–64.

38. Metzger BE. International Association of Diabetes and Pregnancy Study Groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care 2010; 33: 676–82.

39. Farrar D, Simmonds M, Bryant M, Sheldon TA, Tuffnell D, Golder S, et al. Hyperglycaemia and risk of adverse perinatal outcomes: systematic review and meta-analysis. BMJ 2016; 354: i4694.

40. Sweeney AT, Brown FM. Gestational diabetes mellitus. Clin Lab Med 2001; 21: 173–92.

41. Ryan EA. Pregnancy in diabetes. Med Clin North Am 1998; 82: 823–45.

42. American Diabetes Association. Gestational diabetes mellitus. Diabetes Care 1999; 22: S74–S76.

43. Ekelund M, Shaat N, Almgren P, Groop L, Berntorp K. Prediction of postpartum diabetes in women with gestational diabetes. Diabetologia 2010; 53: 452–7.

44. Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U, Coustan DR, Hadden DR, et al. Hyperglycemia and adverse pregnancy outcome (HAPO) study cooperative research group. Hyperglcemia and adverse pregnancy outcomes. N Engl J Med 2008; 358: 1991–2002.

45. American Diabetes Association. Implications of the diabetes control and complications trial. Diabetes Care 1999; 22: S24–S26.

46. Sacks DB, Bruns DE, Goldstein DE, MacLaren K, McDonald JM, Parrott M. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Clin Chem 2002; 48: 436–72.

47. Mosca A, Paleari R, Dalfra MG, Di Canni G, Cuccuru I, Pellegrini G, et al. Reference intervals for hemoglobin A1c in pregnant women: data from Italian multicenter study. Clin Chem 2006; 52: 1138–43.

48. Cornblath M, Schwartz R. Hypoglycemia in the neonate. J Pediatr Endocrinol 1993; 6: 113–29.

49. Silverman BL, Metzger BE, Cho NH, et al. Impaired glucose tolerance in adolescent offspring of diabetic mothers. Diabetes Care 1995; 18: 611–8.

50. Shohat M, Merlob P, Reisner SH. Neonatal polycythemia: I. Early diagnosis and incidence relating to time of sampling. Pediatrics 1984; 73: 7–10.

51. Shohat M, Reisner SH, Mimouni M, Merlob P. Neonatal polycythemia: II. Definition related to time of sampling. Pediatrics 1984; 73: 11–3.

52. Aguilar-Bryan L, Bryan J. Neonatal diabetes mellitus. Endocrine Reviews 2008; 29: 265–91.

53. Ehrmann DA. Polycystic ovary syndrome. N Engl J Med 2005; 352: 1223–36.

54. Nestler JE. Should patients with polycystic ovarian syndrome be treated with metformin? Human Reprod 2002; 17: 1950–3.

55. Kerner W. Klassifikation and Diagnose des Diabetes mellitus. Dtsch Ärztebl 1998; 95: B-2434–8.

56. The NICE-SUGAR Study investigators. Intensive versus conventional glucose control in critically ill patients. N Engl J Med 2009; 360: 1283–97.

57. Cheung BM, Ong KL, Cherny SS, Sham PC, Tso AW, Lam KS. Diabetes prevalence and therapeutic target achievement in the United States, 1999–2006 Am J Med 2009; 122: 443–53.

58. Cryer PE. Mechanisms of hypoglycemia-associated autonomic failure in diabetes. N Engl J Med 2013; 369: 362–72.

59. Juvenile Diabetes Research Foundation. Prolonged nocturnal hypoglycemia is common during 12 months of glucose monitoring in children and adults with type 1 diabetes. Diabetes Care 2010; 33: 1004–8.

60. Agus MSD, Steil GM, Wypij D, Costello JM, Laussen PC, Langer M, et al. Tight glycemic control versus standard care after pediatric cardiac surgery. N Engl J Med 2012; 367: 1208–19.

61. White NH. Diabetic ketoacidosis in children. Endocrinol Metab Clin North Am 2000; 29: 657–82.

62. Delaney MF, Zisman A, Kettyle WM. Diabetic ketoacidosis and hyperglycemic hyperosmolar non ketotic syndrome. Endocrinol Metab Clin North Am 2000; 29: 683–706.

63. Rosenbloom AL. Hyperglycemic hyperosmolar state: an emerging pediatric problem. J Pediatr 2010; 156: 180–4.

64. Skyler JS. Microvascular complications. Retinopathy and nephropathy. Endocrinol Metab Clin North Am 2001; 30: 833–56.

65. American Diabetes Association. Diabetic retinopathy. Diabetes Care 1999; 22: S70–S73.

66. American Diabetes Association. Diabetic nephropathy. Diabetes Care 1999; 22: S66–S69.

67. Rawshani A, Rawshani A, Franzen S, Eliasson B, Svensson AM, Miftaraj M, et al. Mortality and cardiovascular disease in type 1 and type 2 diabetes. N Engl J Med 2017; 376: 1407–18.

68. Giugliano D, Standl E, Visboll T, Betteridge J, Bonadonna R, Campbell IW, et al. Is the current therapeutic armamentarium in diabetes enough to control the epidemic and its consequences? What are the shortcomings. Acta Diabetol 2009; 46: 173–81.

69. Duckworth W, Abraira C, Moritz T, Reda D, Emanuele N, Reaven PD, et al. Glucose control and vascular complications in veterans with type 2 diabetes. N Engl J Med 2009; 360: 129–39.

70. Kelly TN, Bazzano LA, Fonseca VA, Theti TK, Reynolds K, He J. Systematic review: glucose control and cardiovascular disease in type 2 diabetes. Ann Intern Med 2009; 151: 394–403.

71. Capes SE, Hunt D, Malmberg K, et al. Stress hyperglycemia and increased risk of death after myocardial infarction in patients with and without diabetes: A systematic overview. Lancet 2000; 355; 733–8.

72. Malmberg K, Ryden L, Efendic S, et al. Randomized trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute myocardial infarction (DIGAMI study): effects on mortality at 1 year later. J Am Coll Cardiol 1995; 26: 57–65.

73. Boulton AJM, Rayaz AM. Diabetic neuropathy. Med Clin North Am 1998; 82: 909–29.

74. Kuller LH, Dorman JS, Wolf PA. Cerebrovascular disease and diabetes mellitus. In: Harris MI, Hamman RF, eds. Diabetes in America-Diabetes data compiled 1984. US Dept of Health and Human Services. Bethesda 1985; 18: 1–18.

75. Nagi M, Pfefferkorn T, Haberl RL. Blutzucker and Schlaganfall. Nervenarzt 1999; 70: 944–9.

76. Kalaria RN. Diabetes, microvascular pathology and Alzheimer disease. Nature Reviews Neurology 2009; 5: 305–6.

77. Jokl R, Colwell JA. Arterial thrombosis and atherosclerosis in diabetes mellitus. Diabetes Reviews 1997; 5: 1–15.

78. Weir CJ, Murray GD, Dyker AG, et al. Is hyperglycemia an independent predictor of poor outcome of acute stroke? Results of a long-term follow-up study. BMJ 1997; 314: 1303–6.

79. Pomposelli JJ, Baxter JK III, Babineau TJ, et al. Early postoperative glucose control predicts nosocomial infection rate in diabetic patients. J Parenteral and Enteral Nutrition 1998; 22: 77–81.

80. Faulds MH, Dahlman-Wright K. Metabolic diseases and cancer risk. Curr Opin Oncol 2012; 24: 58–61.

81. White Jr JR, Cambell RK. Dangerous and common drug interactions in patients with diabetes mellitus. Endocrinol Metab North Am 2000; 29: 789–902.

82. Gourdy P. Diabetes and oral conception. Clin Endorino l and Metab 2013; 27: 67–76.

83. Vogel A, Strassburg CP, Brabant G, Manns MP. Autoimmun polyglanduläre Syndrome. Dtsch Ärztebl 2002; 99: A1428–A1434.

84. Feig DS, Donovan LE, Corcoy R, Murphy KE, Amiel SA, Hunt KF, et al. Continuous glucose monitoring in pregnant women with type 1 diabetes (Conceptt): a multicentre international randomised controlled trial. Lancet 2017; 390 (10110) 2347–59.

85. Selvin E, Sacks DB. Monitoring glycemic control in endstage renal disease: what should be measured. Clin chem 2017; 63: 447–9.

86. Danne T, Nimri R, Battelino T, Bergenstal RM, Close KC, DeVries JH, et al. International consensus on use of continuous glucose monitoring. Diabetes Care 2017; 40: 1631–40.

3.2 Hypoglycemia syndromes

Lothar Thomas

3.2.1 Definition of hypoglycemia

The term hypoglycemia refers to a low blood glucose concentration associated with clinical symptoms. Hypoglycemia is the result of an imbalance between the inflow of glucose into the bloodstream due to decreased endogenous glucose production or deficient glucose uptake, and the consumption of glucose by the tissues. Hypoglycemia is prevented by a complex regulatory system /1/.

The post absorptive glucose concentration range is 70–100 mg/dL (4.0–5.6 mmol/L). The glucose threshold for a decrease in the blood insulin concentration is approx. 81 mg/dL (4.5 mmol/L). When the glucose level falls to about 65 mg/dL (3.5 mmol/L), there is increased secretion of the counter regulatory hormones glucagon, catecholamines, cortisol, and growth hormone. Glucagon and catecholamines raise the blood glucose level within minutes by stimulating hepatic glycogenolysis and gluconeogenesis as well as renal gluconeogenesis. The substrates of gluconeogenesis are glycerol, free fatty acids, and amino acids. Cortisol and growth hormone reduce the glucose consumption of insulin-sensitive tissues and lead to an increase in blood glucose within hours.

The main source of energy for the brain is glucose, and there are protective mechanisms to maintain glucose homeostasis. When glucose falls to ≤ 57 mg/dL (3.2 mmol/L) in capillary blood /2/ and to ≤ 54 mg/dL (3.0 mmol/L) in venous whole blood /3/, the autonomic (i.e. the sympathoadrenal) nervous system is activated, leading to hypoglycemic symptoms such as anxiety, sweating, tremor, fast heartbeat, and hunger. These end-organ responses, also called autonomic symptoms, can progress to neuroglycopenic symptoms including behavioral changes, cognitive dysfunction, seizures, and coma. However, the threshold for cognitive dysfunction depends on various clinical aspects and psychometric tests. Threshold glycemic levels for symptoms

A study /4/ of 30 healthy subjects who had their capillary blood glucose levels measured 17–18 times per day, showed a daily mean with a standard deviation of 75 ± 14 mg/dL (4.2 ± 0.8 mmol/L). The physiological nadir was reached at 5 PM the level was 70 ± 11 mg/dL (3.9 ± 0.6 mmol/L) and the peak level at 2 PM was 88 ± 18 mg/dL (4.9 ± 1.0 mmol/L). Overall, 5% of glucose levels were below 54 mg/dL (3.0 mmol/L), and 2.8% were below 50 mg/dL (2.8 mmol/L). 33% of participants had levels below 54 mg/dL (3.0 mmol/L), while 17% were below 50 mg/dL (2.8 mmol/L). Since in 95% of cases, blood glucose levels were above 54 mg/dL (3.0 mmol/L), it would make sense to define this concentration as the diagnostic threshold for hypoglycemia /5/. However, consensus statements have defined thresholds of 40 mg/dL (2.2 mmol/L) for venous and capillary whole blood and 50 mg/dL (2.8 mmol/L) for venous plasma in non diabetics /6/. The clinical symptoms associated with a decrease in glucose concentrations are shown in Fig. 3.2-1 – Activation of glucose counter regulatory hormones.

3.2.2 Differentiation of hypoglycemia

According to Whipple /6/, clinical hypoglycemia is present if:

  • There are autonomic and glycopenic symptoms
  • Glucose concentrations are ≤ 40 mg/dL (2.2 mmol/L) in capillary and venous whole blood, and ≤ 50 mg/dL (2.8 mmol/L) in venous plasma.
  • Symptoms resolve after glucose ingestion/administration.

The aforementioned glucose levels are a highly specific criterion for hypoglycemia. According to studies /5/, depending on the specimen (capillary blood, venous blood), even concentrations in the range of 54–63 mg/dL (3.0–3.5 mmol/L) require further investigation if there are clinical symptoms suggestive of hypoglycemia. If levels are below the thresholds suggested by Whipple, further clinical investigations are necessary, even in the absence of hypoglycemia symptoms.

Hypoglycemia is not a diagnosis but a pathological state, the cause of which must be determined. The causes of hypoglycemia are manifold and can be differentiated into:

  • Diabetic hypoglycemia
  • Reactive hypoglycemia
  • Childhood hypoglycemia
  • Hypoglycemia caused by other factors such as physical exercise, alcohol, medications (insulin, salicylates, pentamidine, β-receptor blockers), liver cirrhosis, glucocorticoid deficiency, large tumors, malnutrition, parenteral nutrition, sepsis, shock, insulin antibodies
  • Pseudo hypoglycemia and adrenergic polyprandial syndrome
  • Renal glucosuria
  • Insulinoma
  • Hyperinsulinemic hypoglycemia following gastric bypass surgery. Hypoglycemia syndrome in adults

The most common diagnoses at admission in patients presenting with hypoglycemia are diabetes mellitus, alcoholism, sepsis, and reactive hypoglycemia. Insulinomas are very rare, with a prevalence of 4 cases per 1 million population per year. For evaluation refer to Section 3.7 – Insulin, C-peptide, proinsulin. Iatrogenic hypoglycemia in diabetics is evaluated based on medical history.

To differentiate the hypoglycemia syndrome, patients should be assigned to one of the following categories based on their medical history and clinical presentation /8/:

  • The healthy appearing patient. If the patient has no pre-existing illness, then drug-associated hypoglycemia, alcoholism, reactive hypoglycemia, renal glucosuria, pseudo hypoglycemia, adrenergic postprandial syndrome and insulinoma must be considered primarily. Apparently clinically healthy patients need to undergo intensive laboratory testing for the confirmation and differential diagnosis of hypoglycemia.
  • The ill patient. In this group of patients, hypoglycemia can be associated with the medication of the existing illness (hypertension, diabetes, malaria), or it can be due to a para neoplastic syndrome (non-islet cell tumor hypoglycemia, NICTH), a congenital disorder of carbohydrate metabolism, or an endocrine disorder. Once the diagnosis is known, no further diagnostic evaluation of the hypoglycemia is required.
  • The hospitalized patients who often have serious, multi systemic illnesses. The main causes of the hypoglycemia, apart from diabetes mellitus, are sepsis, shock, liver disease, and renal failure. In these cases, constant blood glucose monitoring is necessary to detect the risk of hypoglycemia.

Hypoglycemia syndromes which are due to an insulinoma predominantly occur in the fasting state, rarely in the fasting plus postprandial state, and very rarely only in the postprandial state.

Postprandial symptoms, which occur 2–4 h after meals are classified as food-stimulated and those which occur more than 5 h after meals are classified as food-deprived. Autonomous symptoms without hypoglycemia, also known as pseudo hypoglycemia, which occur after meals usually cannot confirmed as arising from hypoglycemia. If postprandial hypoglycemia occurs with blood glucose levels below 45 mg/dL (2.5 mmol/L), then the hypoglycemia is stimulated by food intake (e.g., in the case of hereditary fructose intolerance, crop poisoning, or in patients who have had Billroth II surgery) /89/.

The flow chart in Fig. 3.2-2 – Flow chart for the differentiation of hypoglycemia in adults recommends a diagnostic workflow, Tab. 3.2-1 – Hypoglycemia syndromes in adults shows hypoglycemia syndromes in adults. Biomarkers and functional tests for the evaluation of hypoglycemia

Blood glucose: detection of hypoglycemia. If the classic symptoms of hypoglycemia are present, the hypoglycemia etiology is confirmed if at least one of several values is below 45 mg/dL (2.2 mmol/L) in capillary or venous whole blood, and below 50 mg/dL (2.8 mmol/L) in venous plasma, and the remaining values are in the range of 45–54 mg/dL (2.5–3.0 mmol/L). If all values are above 45 mg/dL (2.2 mmol/L) in capillary whole blood, or above 50 mg/dL (2.8 mmol/L) in venous plasma, then hypoglycemia is not confirmed. In this case, the 72-h fast or another functional test should be performed.

In the case of suspected postprandial reactive hypoglycemia, blood glucose self-monitoring is reliable /10/.

The cutoff for the diagnosis of diabetic hypoglycemia is 70 mg/dL (3.9 mmol/L). Findings on hypoglycemia in adults and drug-associated hypoglycemia and their diagnostic significance are listed in Tab. 3.2-1 – Hypoglycemia syndromes in adults.

72-h fast: the test is the mainstay for the evaluation of food-deprived hypoglycemia. Distinction between normal individuals and those with hypoglycemic disorders is based on normal individuals tolerating 3 days food withdrawal without the development of symptoms, whereas patients with hypoglycemic disorder manifest Whipple’s triad usually well short of 72 h /8/. Detection and differentiation of hypoglycemia by determination of insulin, C-peptide and β-hydroxy butyrate (Tab. 3.7-2 – Insulin, C-peptide and proinsulin reference intervals).

C-peptide suppression test, intravenous tolbutamide test, glucagon test: these tests are performed if the 72-h fast is not conclusive. Hypoglycemia in childhood and infancy


The practical threshold for neonates and children is considered to be a blood glucose concentration of 45 mg/dL (2.5 mmol/L). Changes in the cerebral blood flow occur with levels < 30 mg/dL (1.7 mmol/L). All newborns with suspicious clinical symptoms should have their blood glucose maintained at levels > 45 mg/dL (2.5 mmol/L). During the neonatal period, infants of diabetic mothers should have a glucose concentration > 63 mg/dL (3.5 mmol/L), since they have low levels of substrates such as glucose, lactate, alanine and ketone bodies. Every child with a glucose level < 36 mg/dL (2.0 mmol/L) needs to be monitored, even in the absence of clinical symptoms /11/.

Due to the physiological decline in the blood glucose concentration with a nadir 1–2 h after birth, glucose measurements in newborns of nondiabetic mothers should not be performed until 3–4 h after birth, when the physiological hypoglycemia has been overcome /12/. Following enteral feeding, blood glucose levels cycle, with a peak occurring about 1 h after food intake. If hypoglycemia is suspected, a blood sample should be taken just before the second food intake. Low glucose levels in the first 24–48 h are not uncommon in normally developing newborns who are breast-fed.

Infants and adolescents /32/

Every year, approx. 30 in 100,000 children are admitted to hospital with a reduced level of consciousness or non-traumatic coma.

The main etiologies are infections, drug-induced intoxications, seizures, and metabolic disorders. If the condition is not diagnosed timely and correctly, the mortality rate is up to 40%. Tests at hospital admission in children with a reduced consciousness

  • Blood glucose and blood gas analysis.
  • Sodium, potassium, creatinine
  • Urea, ammonia, lactate, ketone bodies
  • AST, LD, GGT, bilirubin
  • Blood count and differential
  • Urinalysis
  • Blood culture.

Freezing of 10 mL urine and 2 mL serum/plasma for possible additional tests.

The possibility of a disease-associated hypoglycemia should be considered if the child has the following history /13/:

  • Short, changing hypoglycemic episodes (2–4 h) not associated with food intake; they occur in hyperinsulinism and type I glycogenosis.
  • Fasting phases of 6–8 h or even 14–16 h are tolerated, but that is not the case in enzyme defect, glycogenolysis or impaired gluconeogenesis
  • Morning fasting hypoglycemia or hypoglycemia during intercurrent illnesses; can be suggestive of impaired gluconeogenesis
  • Postprandial, reactive hypoglycemia; can be suggestive of hereditary fructose intolerance or leucine-sensitive hypoglycemia.

For information on childhood hypoglycemia syndromes refer to:

For the molecular basis of glucose homeostasis and incidence of congenital hypoglycemia see Ref. /33/.


1. Rosen SG, Clutter WE, Berk MA, Shah SD, Cryer PE. Epinephrine supports the post absorptive plasma glucose concentration and prevents hypoglycemia when glucagon secretion is deficient in man. J Clin Invest 1984; 72: 405–11.

2. Mitrakou A, Ryan C, Veneman T, et al. Hierarchy of glycemic thresholds for counter regulatory hormone secretion, symptoms, and cerebral dysfunction. Am J Physiol 1991; 260: E 67–74.

3. Brun JF, Baccara MT, Blacon C, Orsetti A. Niveaux de glycemie veineuse physioligiquement associes aux signes fonctionelles d’hypoglycemie. Comparaison avec des hypoglycemies reactionelles (Abstract). Diabetes Metab 1995; 21 A.

4. Marks V. Glycemic stability in healthy subjects: fluctuations in blood glucose during day. In: Andreani D, Marks V, Lefebvre PJ, eds. Hypoglycemia. New York; Raven Press 1987: 19–24.

5. Brun JF, Fedou C, Mercier J. Postprandial reactive hypoglycemia. Diabetes and Metabolism (Paris) 2000; 26: 337–51.

6. Whipple AO. The surgical therapy of hyperinsulinism. J Internat Chirol 1938; 3: 237.

7. Heller SR. Diabetic hypoglycemia. Bailliaire’s Clinical Endocrinology and Metabolism. 1999; 13: 295–308.

8. Service FJ. Hypoglycemic disorders. N Engl J Med 1995; 332: 1144–52.

9. Comi RJ. Approach to acute hypoglycemia. Endocrinol Metab Clin North Am 1993; 22: 247–62.

10. Palardy J, Havrankova J, Lepage R, Matte R, Belanger R, d’Amour P, Ste Marie LG. Blood glucose measurements during symptomatic episodes in patients with suspected postprandial hypoglycemia. N Engl J Med 1989; 321: 1421–5.

11. Deshpande S, Platt MW. The investigation and management of neonatal hypoglycemia. Semin Fetal & Neonatal Medicine 2005; 10: 351–6.

12. Wendel U. Diagnostisches Vorgehen bei kindlichen Hypoglykämien. Monatsschr Kinderheilkd 1988; 136: 592–6.

13. Cryer PE. Mechanisms of hypoglycemia-associated autonomic failure and its component syndromes in diabetes. Diabetes 2005; 54: 3592–3601.

14. American Diabetes Association Workgroup on Hypoglycemia. Defining and reporting hypoglycemia in diabetes: a report from the American Diabetes Association Workgroup on Hypoglycemia. Diabetes Care 2005; 28: 1245–9.

15. Weitzman ER, Kelemen S, Quinn M, Eggleston EM, Mandl KD. Participatory surveillance of hypoglycemia and harms in an online social network. JAMA Intern Med 2013; 173: 345–51.

16. Service GJ, Thompson GB, Service FJ, Andrews JC, Collazo-Clavell ML, Lloyd RV. Hyperinsulinemic hypoglycemia with nesidioblastosis after gastric-bypass surgery. N Engl J Med 2005; 353: 249–54.

17. Toft-Nielsen M, Madsbad S, Holst JJ. Exaggerated secretion of glucagon-like peptide-1 (GLP-1) could cause reactive hypoglycaemia. Diabetologica 1998; 41: 1180–6.

18. Bergman RN. Toward physiological understanding of glucose tolerance. Minimal model approach. Diabetes 1986; 38: 1512–27.

19. Ahmadpour S, Kabadi UM. Pancreatic alpha-cell function in idiopathic reactive hypoglycemia. Metabolism 1997; 46: 639–43.

20. Sasaki M, Moki T, Wada Y, Hirosawa I, Koizumi A. An endemic condition of biochemical hypoglycemia among male volunteers. Ind Health 1996; 34: 323–33.

21. Marimee TJ, Tyson JE. Hypoglycemia in men. Pathologic and physiologic variants. Diabetes 1977; 26: 161–5.

22. Escalande Polido JM, Alpizar Salazar M. Changes in insulin sensitivity, secretion and glucose effective ness during menstrual cycle. Arch Med Res 1999; 30: 19–22.

23. Zapf J, Futo E, Peter M, Froesch ER. Can big endothelin growth factor II in serum of tumor patients account for the development of extrapancreatic tumor hypoglycemia? J Clin Invest 1992; 90: 2574–84.

24. White Jr JR, Campbell RK. Dangerous and common drug interactions in patients with diabetes mellitus. Endocrinol Metab Clin North Am 2000; 29: 789–802.

25. Bonham JR. The investigation of hypoglycemia during childhood. Ann Clin Biochem 1993; 30: 238–47.

26. Gesellschaft für Neonatologie, pädiatrische Intensivmedizin, et al. Betreuung von Neugeborenen diabetischer Mütter. AWMF-Leitlinie 2010.

27. Roe TF, NG WG, Smit PGA. Disorders of carbohydrate and glycogen metabolism. In: Blau N, Duran M, Blaskovics ME, Gibson KM, eds. Physician’s guide to the laboratory diagnosis of metabolic diseases. Berlin; Springer 2002; 335–55.

28. Duran M. Disorders of mitochondrial fatty acid oxidation and ketone body handling. In: Blau N, Duran M, Blaskovics ME, Gibson KM, eds. Physician’s guide to the laboratory diagnosis of metabolic diseases. Berlin, Springer 2002; 309–34.

29. Birkebaek NH, Simonsen H, Gregersen N. Hypoglycaemia and elevated urine ethylmalonic acid in a child homozygous for the short-chain acyl-CoA dehydrogenase 625G>A gene variation. Acta Paediatr 2002; 91: 480–6.

30. Ryan C, Gurtunca S, Becker D. Hypoglycemia: a complication of diabetes therapy in children. Pediatr Clin N Am 2005; 1705–33.

31. Diabetes Control and and Complications Trial Research Group. Hypoglycemia in the Diabetes Control and Complications Trial. Diabetes 1997; 46: 271–86.

32. Bowker R, Green A, Bonham JR. Guidelines for the investigation and management of reduced level of consciousness in children: implications for clinical biochemistry laboratories. Ann Clin Biochem 2007; 44: 506–11.

33. Lang TF. Update on investigating hypoglycemia in childhood. Ann Clin Biochem 2011; 48: 200–11.

3.3 Blood glucose

Lothar Thomas

Conventional blood glucose should be determined in capillary whole blood or venous plasma /1/. With other types of samples, different glucose concentrations are measured at the same sampling time in the same individual /2/. This must be taken into account during the clinical evaluation.

A blood glucose test is a measure of glucose concentration present in an the blood of an individual at a given point of time. Diagnostic laboratory tests for diabetes are (Tab. 3.3-1 – Blood glucose specimen for evaluation of the metabolic state):

  • Fasting glucose (FG), an excellent test for "in moment glucose"
  • 2-hour glucose during a 75-g oral glucose tolerance test (oGGT) for diagnosis of impaired glucose tolerance
  • Random glucose (measurement of glucose at not given point)
  • Continuous glucose monitoring using a chip implanted under the skin.

3.3.1 Indication

Suspected hyperglycemia

  • Screening for diabetes mellitus in ambulant patient care and hospitals
  • Monitoring of diabetes therapy
  • Evaluation of carbohydrate metabolism (e.g., pregnant women, patients with obesity, hyperlipidemia, cardiovascular disease, stroke, renal failure, patients with reduced consciousness or in a coma, patients with liver disease, acute hepatitis, acute pancreatitis, chronic pancreatopathy, autoimmune polyglandular syndrome, acromegaly, Addison’s disease, panhypopituitarism, therapy with corticosteroids and drugs that induce hyperglycemia, stress response).

Suspected hypoglycemia

  • Diabetes therapy and occurrence of hypoglycemia symptoms
  • Exclusion of hypoglycemia syndrome in clinically apparently healthy individuals (exclusion of insulinoma)
  • Hypoglycemia symptoms in the critically ill
  • Diagnosis of neonatal hypoglycemia
  • Suspected congenital metabolic disorder
  • Treatment with drugs that induce hypoglycemia.

3.3.2 Method of determination

There are different methods for the determination of glucose in blood and body fluids /1/.

Glucose oxidase method

Principle: the enzyme glucose oxidase catalyzes the oxidation of glucose to gluconic acid and H2O2. In the subsequent peroxidase-mediated indicator reaction, H2O2 oxidizes a reduced chromogen to produce a colored compound, which is measured using a photometer. The color intensity of the oxidized chromogen is proportional to the glucose concentration /3/.

α-D-glucose spontaneous β-D-glucose β-D-glucose + H 2 O 2 + O 2 reduced chromogen + H 2 O 2 peroxidase oxid. chromogen + 2 H 2 O glucose oxidase D-gluconolactone + H 2 O 2

Hexokinase method /4/

Principle: hexokinase in the presence of ATP phosphorylates glucose to form glucose-6-phosphate. The latter reacts with NADP to form 6-phosphogluconate and NADPH2. This reaction is catalyzed by glucose-6-phosphate dehydrogenase (G-6-PD). The measurand is NADPH2, the increase in NADPH2 is measured at the endpoint of the reaction. The increase in absorbance determined is proportional to the glucose concentration in the test sample.

D-glucose + ATP hexokinase D-glucose-6-phosphate + ADP Mg 2+ D-glucose-6-phosphate + NADP + G-6-PDH D-glucose-6-phosphate + NADP + H +

Glucose dehydrogenase (Gluc-DH) method /5/

Principle: glucose is oxidized to gluconolactone by Gluc-DH. The hydrogen released in the reaction is transferred to NAD, producing NADH2. The increase in NADH2 is measured using the principle of continuous absorbance registration. The increase in absorbance is proportional to the glucose concentration in the test sample. In contrast to the end point method addition of mutarotase to the reagents is not necessary.

α-D-glucose Mutarotase β-D-glucose β-D-glucose + NAD + Gluc-DH D-gluconolactone + NADH + H +

Gluc-DH only reduces β-D-glucose. In aqueous solution, glucose is present in the α- and β-form. As the β-D-glucose is consumed, an equilibrium between the two forms is established again as a function of time. To prevent this reaction from becoming the determining factor for the speed of the Gluc-DH reaction, the reagent contains mutarotase. This enzyme accelerates the rate at which equilibrium is reached.

Measurement with a biosensor /6/

Biosensors are analytical devices that incorporate a biological material (e.g., the enzyme glucose oxidase) and are connected to an optical or electrochemical detection system.

Principle of the glucose sensor: in the first step, glucose reacts with the oxidized form of the enzyme glucose oxidase (GOD) to form gluconic acid. In this process, two electrons and two protons are released, and GOD is reduced. In the second step, O2 which is present in the surrounding fluid reacts with GOD accepting the aforementioned electrons and protons leading to form H2O2 and regenerating oxidized GOD, which is ready to react once more with glucose. The glucose concentration in the test sample determines the amount of H2O2. This is detected following oxidation at the surface of a platinum electrode which causes a change in the electrochemical potential.

Measurement with glucose meters

Analyzers in which glucose is determined using readable strip and reflectance photometer are used for:

  • Point-of-care testing in intensive care units in hospitals or outpatient clinics, in facilities for chronic disease management, and in doctor’s office
  • Self-monitoring of blood glucose (SMBG) at home, at work, or at school. In the USA, national standards were developed for SMBG /8/.

Glucose tests with glucose meters are based on the photometric measurement of the color development of a chromogen or on the principle of the glucose electrode /9/. With the photometric measurement, glucose is enzymatically oxidized to gluconolactone by the enzymes glucose peroxidase or glucose dehydrogenase.

In the subsequent indicator reactions,

  • the H2O2 produced, catalyzed by peroxidase, oxidizes the 3,3’, 5,5’ tetra methyl benzidine to a blue dye whose intensity is measured with a reflectometer,


  • the NADH produced, catalyzed by diaphorase, reduces the dye 3-(4’,4’-dimethylthiazole-2-yl)-2,4-diphenyl tetrazolium bromide to a formazan dye.

The optimal sample is capillary blood. Modern glucose meters for the self-monitoring of blood glucose allow the storage and processing of the measured values and the calculation of mean blood glucose (MBG) and mean amplitude of glucose excursions (MAGE).

Continuous glucose monitoring /10/

A small chip is implanted under the skin, which provides continuous glucose monitoring readings from the interstitial fluid to the sensor kept outside. Invasive glucose sensors use enzyme electrodes or micro dialysis systems for sensing of glucose. Micro dialysis systems use a fine, hollow micro dialysis fibre placed subcutaneously. Non-invasive glucose sensors use optical transducers. The transducers use light in variable frequencies to detect glucose.

3.3.3 Specimen

Depending on the method of determination, the following are used:

  • Capillary blood, depending on the sampling procedure: 0.01–0.02 mL
  • Venous (rarely capillary) plasma: 0.01–0.05 mL

3.3.4 Reference interval

Refer to Tab. 3.3-2 – Blood glucose reference intervals.

3.3.5 Clinical significance

The blood level of glucose depends on the metabolic state of an individual. The following states are possible (Tab. 3.3-1 – Blood glucose specimen as a function of the metabolic state):

  • Post absorptive state; period of 6–12 h after the start of food intake. During this period the transition from the postprandial to the fasting state occurs.
  • Postprandial state; this comprises a period of 2–3 h after the start of food intake. During this period, the blood glucose level in plasma rises up to 200 mg/dL (11.1 mmol/L) beginning 10 min. after the start of a meal. Even though glucose levels return to preprandial levels within 2–3 h after food intake, it takes about 6 h for a meal to be completely assimilated and for the post absorptive state to be restored. Compared to the fasting state, there is hyperglycemia. This depends on the type of carbohydrates consumed, the amount of fat and proteins, the size of the meal, and the time of day. Postprandial glucose accounts for approx. 30–40% of the total daytime hyperglycemia.
  • Fasting state; comprises a period of 8–10 h after the last food intake. The glucose concentration of non diabetics is below 100 mg/dL (5.6 mmol/L).

The behavior of blood glucose in different conditions is shown in:

and the diagnostic significance of glucose in the fasting and postprandial state in:

Glucose levels within the reference interval do not rule out diabetes, and concentrations above the reference interval do not confirm it /13/. This is due to intraindividual variations of blood glucose levels which are greater than those of other blood parameters as they are influenced by physical activity and the length of time since the last food intake. For example, in healthy individuals the intraindividual (biological) and inter individual variation of fasting capillary glucose [mean glucose of 88 mg/dL (4.9 mmol/L)] is 4.8–6.1% and 7.5–7.8% respectively /1415/. The biological variability of plasma glucose is thus higher than the analytical imprecision. Moreover, the fasting plasma glucose concentration increases continuously with age from the third to the sixth decade of life. Dysregulations such as insulin resistance, hyperinsulinism and diabetes as well as pregnancy further increase the variations. In newly diagnosed type 2 diabetics, the intraindividual variation of fasting glucose is 13.7%, and the inter individual variation 14.8% /16/. The interpretation of blood glucose levels also depends on the type of sample examined.

3.3.6 Comments and problems

Sample materials

The concentration of glucose in blood depends on the type of sample examined. Due to the higher water content compared to the red blood cells, glucose concentration measured in venous plasma is generally 10–18% higher than in venous whole blood. Arterial whole blood has a higher glucose concentration than venous blood; the glucose concentration of capillary whole blood sampled from the finger tip is in between the two. Measurements in capillary whole blood and venous plasma result in similar glucose levels within the reference interval.

The following types of specimen are used in the different countries for determining blood glucose in routine diagnosis: Capillary whole blood, venous whole blood, and plasma from venous whole blood.

Capillary whole blood: samples should be collected by skin puncture from the finger or from the heel (infants only). In the fasting state there is no arteriovenous difference between arterial and venous blood. Therefore, the concentrations measured in venous and capillary whole blood are nearly identical. In the postprandial state, however, there may be a 20–70% difference /26/. The arteriovenous difference is greatest in lean nondiabetic individuals, smallest in diabetics, and larger with blood sampled from deep veins compared to blood from superficial veins /1/.

Compared to glucose measured in plasma, the glucose concentration in whole blood is influenced by the hematocrit (Hct), by proteins, lipoproteins and other dissolved and corpuscular components. The corresponding values for a glucose concentration in water of 180 mg/dL (10.0 mmol/L) are as follows /1/:

  • In plasma: 168 mg/dL (9.3 mmol/L).
  • In whole blood with 0.30 Hct: 155 mg/dL (8.6 mmol/L).
  • In whole blood with 0.45 Hct: 150 mg/dL (8.3 mmol/L).
  • In whole blood with 0.60 Hct: 144 mg/dL (8.0 mmol/L).

Venous plasma: the molality of glucose in whole blood and plasma is identical. However, the volume of water is about 11% higher in plasma than in whole blood. Therefore, glucose levels are also about 11% higher in plasma than in whole blood at a hematocrit of 0.43.

Serum: glucose levels are 5% lower in serum than in heparinized plasma /27/. Glucose levels in different types of sample

A study /28/ has established the following conversion factors for the sample types capillary whole blood, venous whole blood, and venous plasma:

  • Venous plasma/venous whole blood in diabetics and non diabetics: 1.148.
  • Venous plasma/capillary whole blood in non diabetics: 0.997.
  • Venous plasma/capillary whole blood in diabetics: 1.089 and overall mean 1.048.
  • Capillary whole blood/venous whole blood in non diabetics: 1.173.
  • Capillary whole blood/venous whole blood in diabetics: 1.055 and overall mean 1.155.


The International Federation of Clinical Chemistry (IFCC) recommends reporting the concentration of glucose in plasma, irrespective of the sample type and assay /29/. A constant factor of 1.11 is used to convert concentration in whole blood to the equivalent concentration in plasma. This applies to a Hct of 0.43. In the case of higher Hct values, as are typical in neonates, this factor must be increased by multiplication by the following correction factor (cf):

cf = 0.84/(0.93–0.22 Hct)

With a Hct of 0.70 and multiplication of 1.11 with cf, the new factor to convert concentration in whole blood to the equivalent concentration in plasma is 1.19.

According to the IFCC, it is possible to convert whole blood glucose and biosensor glucose to plasma glucose, but not whole blood glucose to biosensor glucose (Fig. 3.3-1 – Conversion factors for glucose). According to recommendations of the WHO, the cutoffs for fasting glucose and for the oral glucose tolerance test are identical for capillary whole blood and venous plasma.

Blood sampling

Fasting glucose: sampling 7 a.m. to 8 a.m. after at least 8 h of fasting.

Postprandial glucose: 1–2 h after a meal.

Type of sample

Capillary whole blood: only draw blood if blood circulation is good; finger must be warm. 0.01–0.02 mL of blood is added to hemolyzing solution.

Venous blood: is analyzed in the form of whole blood, plasma, and serum. In plasma and serum following blood collection is recommended in separator tubes. The collection tubes for determining glucose in whole blood contain NaF to prevent glycolysis, and potassium oxalate or Na2EDTA to inhibit clotting. NaF acts by inhibiting glycolytic enzymes, in particular enolase, although the effect is minor in the first 2 h after blood collection. A better effect than with NaF alone is achieved by cooling the sample, by acidifying it, or by using citrate tubes for blood collection. If both glucose and lactate levels are to be determined, collection tubes containing NaF and citrate are suited best /30/.

Method of determination

The reference method is the hexokinase method, or in some countries the glucose oxidase method. Possible methodological errors are shown in Tab. 3.3-5 – Methodological errors in glucose measurement.


In the glucose oxidase method, Novaminsulfon (metamizole) and ascorbic acid in concentrations > 0.4 g/l and α-methyldopa > 0.2 g/l can cause a decrease in the glucose level of up to 50% /31/.

Icodextrin /32/: the glucose polymer icodextrin is often added to the dialysis fluid during peritoneal dialysis so that an osmotic gradient can be maintained along the dialysis membrane and the ultrafiltration time can be prolonged. However, small amounts of icodextrin can get into the bloodstream via the lymphatic system. In the bloodstream it is hydrolyzed to glucose oligomers such as maltose and maltotriose. These oligomers cause falsely high glucose readings in some point-of-care glucometers.


Due to glycolysis, the glucose concentration in whole blood decreases by 5–7% (approx. 10 mg/dL; 0.6 mmol/L) per hour after blood sampling. At 4 °C there is only a slight decrease during the first 2 h and approx. 20% after 24 h /33/. The decrease depends on the glucose concentration, the ambient temperature, and the leukocyte count /13/. The decrease in glucose in whole blood within the first 2 h after blood sampling is approximately the same with and without NaF /34/.

In EDTA-coated collection tubes there is no significant decrease within 24 h in the presence of maleinimide.

Glucose concentrations measured in the serum/plasma of blood collected in tubes containing separator gel were comparable to those measured in plasma collected in tubes containing NaF and potassium oxalate /35/.

At 4 °C blood deproteinized by perchloric acid gives stable values in the supernatant, obtained by centrifugation, for at least 5 days.

Capillary blood, stabilized in the mentioned hemolyzation solutions shows stable glucose values for 48 h /2/.

In newborns, measurement of blood glucose should be performed as soon as possible after blood collection, since the rate of glycolysis of erythrocytes in newborns is considerably higher than in adults so that the glycolysis inhibitors cannot be as effective. Some cases of neonatal hypoglycemia are reported to be falsely low, particularly in those newborns with high Hct values /36/.


1. Marks V. Blood glucose: its measurement and clinical importance. Clin Chim Acta 1996; 251: 3–17.

2. Weiner K. Whole blood glucose. What are we actually measuring. Ann Clin Biochem 1995; 32: 1–8.

3. Reljic R, Ries M, Anic N, Ries B. New chromogen for assay of glucose in serum. Clin Chem 1992; 38: 522–5.

4. Kunst A, Draeger B, Ziegenhorn J. UV-methods with hexokinase and glucose-6-phosphate dehydrogenase. In: Bergmeyer HU, ed. Methods of enzymatic analysis. Weinheim: Verlag Chemie, 1984: Vol IV, 163–72.

5. Vormbrock R. UV-method with glucose dehydrogenase. In: Bergmeyer HU, ed. Methods of enzymatic analysis. Weinheim: Verlag Chemie, 1984: Vol IV, 172–8.

6. Turner APF, Chen B, Piletschy SA. In vitro diagnostics in diabetes: meeting the challenge. Clin Chem 1999; 45: 1596–1001.

7. Monnier L. Is postprandial glucose a neclected cardiovascular risk factor in type 2 diabetes? Eur J Clin Invest 2000; 30, Suppl 2: 3–11.

8. Mensing C, Boucher J, Cypress M, Weinger K, Mulcahy K, Barta P, et al. National standards for diabetes self-management education. Diabetes Care 2005; 28, Suppl 1: S72–9.

9. Schlebusch H. Dezentrale Blutglucosebestimmungen im Krankenhaus. DG Klin Chem Mitt 1996; 27: 91–103.

10. Oliver NS, Toumazou C, Cass AEG, Johnston DG. Glucose sensors: a review of current and emerging technology. Diabetic Medicine 2009; 26: 197–210.

11. Heck LJ, Erenberg A. Serum glucose levels in term neonates during the first 48 hours of life. J Pediatr 1987; 110: 119–22.

12. American Diabetes Association. Standards of medical care in diabetes – 2014. Diabetes Care 2014; 37, Suppl 1: S14–S80.

13. Sacks DB, Bruns DE, Goldstein DE, Maclaren NK, McDonald JM, Parrott M. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Clin Chem 2002; 48: 436–72.

14. Widjaja A, Morris RJ, Levy JC, Frayn KN, Manley SE, Turner RC. Within and between subject variation in commonly measured anthropometric and biological variables. Clin Chem 1999; 45: 561–6.

15. Sebastian-Gambano MA, Liron-Hernandez FJ, Fuentes Arderiu X. Intra- and inter-individual biological variability data bank. Eur J Clin Chem Clin Biochem 1997; 35: 845–52.

16. Ollerton RL, Playle R, Ahmed K, Dunstan FD, Luzio SD, Owens DR. Day-to-day variability of fasting glucose in newly diagnosed type 2 diabetic subjects. Diabetes Care 1999; 22: 394–8. 12

17. American Diabetes Association. Standards of medical care in diabetes. Diabetes Care 2005; 35, Suppl 1: S11–63.

18. Ceriello A. Acute hyperglyemia: a new risk factor during myocardial infarction. Eur Heart J 2005; 26: 328–31.

19. Els T, Klisch J, Orszagh M, Hetzel A, Schulte-Mönting J, Schumacher M, Lücking CH. Hyperglycemia in patients with focal cerebral ischemia after intravenous thrombolysis: influence on clinical outcome and infarct size. Cerebrovasc Dis 2002; 13: 89–94.

20. Kagansky N, Levy S, Knobler H. The role of hyperglycemia in acute stroke. Arch Neurol 2001; 58: 1209–12.

21. Lukins MB, Manninen PH. Hyperglycemia in patients administered dexamethasone for craniotomy. Anesth Analg 2005; 100: 1129–33.

22. Crane PK, Walker R, Hubbard RA, Li G, Nathan DM, Zjeng H, et al. Glucose levels and risk of dementia. N Engl J Med 2013; 369: 540–8.

23. Behrendt CE, Ruiz RB. Hyperglycemia among persons with hepatitis C: not the classical diabetic phenotype. Diabetes Research & Clinical practice 2006; 71: 68–74.

24. Pasquel FJ, Spiegelman R, McCauley M, Smiley D, Umpierrez D, Johnson R, et al. Hyperglycemia during total parenteral nutrition. Diabetes Care 2010; 33: 739–41.

25. Hey E. Hyperglycemia and the very preterm baby. Semin Fetal Neonatal Med 2005; 10: 377–87.

26. Duffy T, Phillips N, Pellegrin F. Review of glucose tolerance: a problem in methodology. Am J Med Sci 1973; 265: 117–33.

27. Ladenson JH, Tsai LM, Michael JM, Kessler G, Joist JH. Serum versus heparinized plasma for eighteen common chemistry tests: is serum the appropriate specimen? Am J Clin Pathol 1974; 62: 545–52.

28. Haeckel R, Brinck U, Colic D,Janka HU,Püntmann I, Schneider J, Vierbrock C. Comparability of blood glucose concentrations measured in different sample systems for detecting glucose intolerance. Clin Chem 2002; 48: 936–9.

29. D’Orazio P, Burnett RW, Fogh-Andersen N, Jacobs E, Kuwa K, Külpmann WR, et al. Approved IFCC recommendation on reporting results for blood glucose Clin Chem Lab Med 2006; 44: 1486–90.

30. Bruns DE. Are fluoride-containing blood tubes still needed for glucose testing? Clin Biochem 20131; 46: 289–90.

31. Szasz G, Huth K, Busch EW, Koller PU, Stähler F, Vollmar J. In vivo drug interference with various glucose determinations compared with in vivo results. Z Klin Chem Klin Biochem 1974; 12: 256–61.

32. Apperloo JJ, Vader HL. A quantitative appraisal of interference by icodextrin metabolites in point-of-care glucose analysis. Clin Chem Lab Med 2005; 43: 314–8.

33. Miller SA, Wallace RJ, Musker DM, Septimus EJ, Kohl S, Baughn R. Hypoglycemia as a manifestation of sepsis. Am J Med 1980; 68: 649–54.

34. Mikesh LM, Bruns DE. Stabilization of glucose in blood specimens: mechanisms of delay in fluoride inhibition of glycolysis. Clin Chem 2008; 54: 930–2.

35. Chan AYW, Ho CS, Cockram CS, Swaminathan R. Handling of blood specimens for blood glucose analysis. J Clin Chem Clin Biochem 1990; 28: 185–6.

36. Fernandez L, Jee P, Klein MJ, Fischer P, Brooks SPJ. A comparison of glucose concentration in paired specimens collected in serum separator and fluoride/potassium oxalate blood containing tubes under survey field conditions. Clin Biochem 2013; 46: 285–8.

37. Tate PF, Clemens CA, Walters JE. Accuracy of home blood glucose monitoring. Diabetes Care; 1992; 15: 536–8.

38. Kane MJO, Pickup J. Self-monitoring of blood glucose in diabetes: is it worth it? Ann Clin Biochem 2009; 46: 2730–82.

3.4 Glucose in urine and extravascular fluids

Lothar Thomas

3.4.1 Indication

Urinary glucose

  • Detection of glucosuria of unknown etiology
  • Monitoring of diabetes mellitus therapy in patients who are unable to self-monitor their blood glucose levels.

Glucose in cerebrospinal fluid: suspected bacterial meningitis.

Glucose in extravascular fluids: suspected bacterial infection.

3.4.2 Method of determination

Qualitative, semi quantitative determination in urine

Test strip methods /1/: these methods are based on a glucose oxidase/peroxidase reaction with tetra methyl benzidine as redox indicator. The color of the reagent pad changes from yellow to green with increasing glucose concentration in the sample.

Other test strips use a potassium iodide chromogen instead of tetra methyl benzidine. In this case, the peroxidase-catalyzed oxidation causes the chromogen to turn from green to brown with increasing glucose concentration.

The semiquantitative tests work by the same principle of reaction. The intensity of the color that develops on the test pad indicates the glucose concentration in g/l.

Quantitative determination in urine

See blood glucose in Section 3.3 – Blood glucose.

3.4.3 Specimen

First- or second-void urine or urine of defined sampling periods, indication of collection volume, supplementation with 1 g of sodium azide per 24 h sampling period.

Other body fluids such as ascites: 0.1–1 mL

3.4.4 Reference interval

Refer to Tab. 3.4-1 – Glucose reference intervals in urine and body fluids.

3.4.5 Clinical significance Urinary glucose and diabetes mellitus

The extent of glucosuria is the result of the glomerular filtration and tubular reabsorption of glucose. Up to a blood glucose concentration of 160–180 mg/dL (8.9–10.0 mmol/L), also called the renal threshold, all filtered glucose is reabsorbed by the renal tubules. When the blood glucose level exceeds the renal threshold, glucosuria occurs, which is an indirect indicator of hyperglycemia (Tab. 3.4-2 – Diseases and conditions associated with glucosuria).

Therefore, the finding of glucosuria is suggestive of the presence of diabetes and always requires further investigation. The urine test strip is unsuitable as a method of screening for diabetes. According to a study /3/, the diagnostic sensitivity is 55% with a specificity of 99%, the positive predictive value is 29% and the negative predictive value is 95%. This is due to the fact that in diabetics, especially elderly people, the renal threshold is raised.

Urinary glucose is no longer of relevance in the monitoring of glucose control in diabetics, because it is only a rough indicator of the glycemic state/4/. On the one hand, its usefulness is limited by the renal threshold of about 180 mg/dL (10.0 mmol/L), on the other hand a certain correlation between blood glucose concentration and urinary glucose excretion exists only with blood glucose levels up to about 140 mg/dL (7.8 mmol/L) (Fig. 3.4-1 – Urinary glucose excretion as a function of blood glucose concentration/5/.

Pre-term infants are given glucose infusions for nutritional purposes. The amount of glucose administered should be limited such that the glucose concentration in urine does not exceed 20 g/l (see also Tab. 3.3-3 – Blood glucose in diabetes and various conditions). Renal glucosuria

Renal glucosuria associated with normoglycemia /5/:

  • Glucose-phosphate diabetes (glucosuria and phosphaturia)
  • Fanconi syndrome (glucosuria, phosphaturia, aminoaciduria)
  • Acquired tubular injury (pyelonephritis, glomerulonephritis, intoxication)
  • Pregnancy
  • Renal diabetes.

Renal diabetes is the most common of the various forms of glucosuria with normoglycemia.

Renal diabetes

Renal diabetes is a dominantly inherited form of glucosuria which mainly occurs in men. It is due to diminished reabsorption of glucose in the proximal tubules. In healthy individuals glucose is nearly completely reabsorbed in the proximal tubules and with a daily mean blood glucose level of 100 mg/dL (5.6 mmol/L) and a glomerular filtrate of 125 [mL × min.–1 × (1,73 m2)–1] 180 g of glucose is reabsorbed daily. Fractional glucose extraction

Renal glucosuria is characterized by determining its fractional glucose extraction (FEG). This is the ratio of the glucose excreted in urine and the glomerular filtered glucose. The FEG is determined in first-void morning urine and calculated using the following equation:

FE G (%) = Urinary G (mg/dL) × Serum Cr (mg/dL) Serum G (mg/dL) × Urinary Cr (mg/dL)

Cr, creatinine; G, glucose

The reference interval of FEG (%) =(1.5–7.5) × 10–4. In renal glucosuria, the FEG is generally decreased. In gestational glucosuria it is on average 22.9 × 10–4 /6/.

Diagnostic information about glucose in cerebrospinal fluid and other extravascular body fluids are shown in Tab. 3.4-3 – Diagnostic significance of glucose measurements in CSF and extravascular fluids.

3.4.6 Comments and problems

Method of determination

The lower detection limit of the glucose test strip methods is 30–50 mg/dL (1.7–2.8 mmol/L). Concentrations over 250 mg/dL (13.9 mmol/L) are measured with increased imprecision.

The hexokinase and glucose dehydrogenase methods have no interference from substances physiologically occurring in urine, and negligible interference from drugs. Interferences with the use of reagent strips are shown in Tab. 3.4-4 – Interferences in urine glucose determination using the glucose oxidase method.

Both methods can be used for quantitative determination of glucose in urine. The hexokinase method is used to detect elevated levels in the rare condition of fructosuria. Fructose excretion is also increased if high doses of fructose (e.g., diabetic sweets) are ingested orally.

Stability in urine

The measurement should be performed within 2 h after urine collection. Unless the urine contains stabilizing additives, approx. 40% of glucose is lost within 24 h /11/. In the presence of bacteriuria, leukocyturia or hematuria there is an even greater decrease in glucose levels. To stabilize the urine sample, it is recommended to add sodium azide so that its final concentration in urine is approximately 1%.


1. Smalley DL, Bradley ME. New test for urinary glucose (BM 33071) evaluated. Clin Chem 1985; 31: 90–2.

2. Heimsoth VH, Graffe-Achelis Ch, Banauch D, Vollmar J. Referenzwerte für die Glucosekonzentration im Urin von Erwachsenen. Med Labor 1978; 31: 236–40.

3. Mehnert H, Sewering H, Reichstein W, Vogt H. Früherfassung von Diabetikern in München. Dtsch Med Wschr 1968; 93: 2044–7.

4. American Diabetes Association. Tests of glycemia in diabetes. Diabetes Care 1999; 22: S77–S79.

5. Küntzel W, Mitzkat HJ. Zur Diagnostik der renalen Glucosurie. Dtsch Med Wschr 1971; 96: 1130–2.

6. Renschler H. Der Einfluss der Nierenfunktion auf die Glucoseausscheidung Gesunder. Habilitationsschrift; Heidelberg 1964.

7. Kornelisse RF, de Groot R, Neijens HJ. Bacterial meningitis: mechanisms of disease and therapy. Eur J Pediatr 1995; 154: 85–96.

8. Runyon BA, Hoefs JC, Morgan TR. Ascites fluid analysis in malignancy related ascites. Hepatology 1988; 8: 1104–9.

9. Boggs DS, Kinasewitz GT. Review: pathophysiology of the pleural space. Am J Med Sci 1995; 309: 53–9.

10. Berg B. Ascorbate interference in the estimation of urinary glucose by test strips. J Clin Chem Clin Biochem 1986; 24: 89–96.

11. Lodd JA, Turner K. Evaluation of Trinder’s glucose oxidase method for measuring glucose in serum and urine. Clin Chem 1975; 21: 1754–6.

3.5 Oral glucose tolerance test (oGTT)

Lothar Thomas

The oGTT is the standard test for the diagnosis of impaired glucose tolerance (IGT) and describes the postprandial glucose state. In prediabetes and diabetes mellitus glucose tolerance is impaired. The American Diabetes Association (ADA) recommends fasting plasma glucose (FPG) as an acceptable screening test for prediabetes and diabetes and classifies an elevated FPG level as impaired fasting glucose (IFG) /1/. The oGTT is recommended for confirming IGT in patients with IFG. In Europe the oGTT is the preferred screening test for prediabetes and diabetes as it allows the detection of both IFG and IGT in a single test /2/.

3.5.1 Indication

The ADA recommends the performance of an oGTT in /345/:

  • Individuals with IFG (100–125 mg/dL; 5.6–6.9 mmol/L)
  • Individuals ≥ 45 years
  • All individuals, regardless of age, with a body mass index (BMI) ≥ 25 kg/m2 and at least one additional risk factor. Risk factors include: first-degree relatives with type 2 diabetes (T2D), arterial hypertension, dyslipidemia, cardiovascular disease, history of gestational diabetes, member of an ethnic group with a high prevalence of diabetes (see also Tab. 3.1-2 – Diagnosis of prediabetes and diabetes mellitus based on ADA criteria).
  • In pregnant women at 24–28 weeks of gestation
  • Individuals with glucosuria and normal FPG.

3.5.2 Test procedure

Preparation of the patient

To oGTT will provide valuable results if the following requirements have been met by the patient prior to the test:

  • No caloric intake for at least 10–16 h.
  • Maintenance of usual eating habits for at least 3 days (≥ 150 g carbohydrates per day)
  • Dis continuation of interfering medications for at least 3 days prior to the test if this is possible without risk to the patient
  • Test should be performed with the patient seated or lying down (no muscular effort); no smoking prior to or during the test
  • A time interval to menstruation of at least 3 days.

Performance of the oGTT

After the collection of capillary or venous blood samples to determine fasting glucose, the patient drinks the following solutions, dissolved in 250–300 mL of water, over 5 minutes:

  • 75 g of water-free glucose, or
  • 82.5 g of glucose monohydrate, or
  • an isocaloric amount of hydrolyzed starch.

Children are given 1.75 g per kg of body weight, up to a maximum of 75 g.

The test begins in the morning, between 8 a.m. and 9 a.m. After a fasting blood sample has been collected and the individual has started to drink the glucose solution, another blood sample is taken after 120 min. During the test, the patient should be at rest without stress.

Perfomance of oGTT for gestational diabetes

in Europe the 75-g oGTT is performed. Blood samples are collected at fasting as well as 60 and 120 min. after the start of the glucose drink.

In North America diagnosis of gestational diabetes can be accomplished with either of two strategies:

  • One-step 75 g oGTT or
  • Two-step approach with a 50-g (non-fasting) screen followed by a 100-g oGTT for those who screen positive.

3.5.3 Specimen

Capillary or venous blood for determining blood glucose, per sample: 0.01–0.02 mL

3.5.4 Clinical significance Diagnosis of prediabetes and diabetes

The oGTT combines the measurements of fasting plasma glucose (FPG) and postprandial glucose (2-h level). The FPG value provides information about insulin secretion while the 2-h value is an indicator of insulin resistance. The following results are possible (Tab. 3.5-1 – Diagnosis of diabetes mellitus and impaired glucose tolerance using 75 g oGTT):

  • Elevated FPG with normal 2-h value, the condition of impaired fasting glucose (IFG).
  • Normal FPG with elevated 2-h value, the condition of impaired glucose tolerance (IGT).
  • Combination of IFG and IGT without reaching diabetic glucose levels; a condition termed IFG/IGT.

In contrast to the ADA, the European Association for the Study of Diabetes prefers to recommend the oGTT for diabetes screening. The ADA recommends performing the oGTT after the FPG or the HbA1c in cases where the risk of diabetes needs to be better differentiated /3/.

Since the oGTT measures both IFG and IGT, it identifies more individuals at risk of developing T2D than the IFG test.

To diagnose gestational diabetes in pregnant women, the ADA alternatively recommends /3/:

  • The two-step approach, in which the 50-g oGTT is performed initially, followed by the 100-g oGTT for women who screen positive on the initial test
  • The one-step approach, in which the 75-g oGTT is performed primarily. Blood samples are collected at fasting as well as 60 and 120 min. after the start of the glucose drink.

When interpreting the glucose results of the oGTT, the influencing factors listed in Tab. 3.5-2 – Screening for and diagnosis of gestational diabetes using 75 g oGTT must be taken into consideration. Fasting plasma glucose (FPG)

The FPG value provides information on whether insulin secretion is diminished or there is increased hepatic glucose production /8/. If insulin secretion is diminished, hepatic glucose production in the post absorptive state will be increased, resulting in hyperglycemia. Individuals with normal FPG (below 100 mg/dL; 5.6 mmol/L) only have a 5.5% risk of developing T2D, while the risk is much higher in individuals with concentrations ≥ 126 mg/dL (7.0 mmol/L). The result of the oGTT is of importance in individuals with FPG levels in the range of 100–109 mg/dL (5.6–6.0 mmol/L) and 110–125 mg/dL (6.1–6.9 mmol/L), since there is a 9% respectively 26% chance that the oGTT result is diagnostic of diabetes /9/. 2-h postprandial glucose value

The 2-h glucose value from the oGTT is an indicator of impaired glucose tolerance (IGT) and results from the insulin resistance of the tissues, in particular muscle, liver and fat tissues /8/. The glucose load stimulates insulin secretion. This does, however, not lead to glucose absorption by the tissues, since the tissue insulin receptors do not respond adequately. As a result, there is prolonged hyperglycemia due to reduced glucose clearance. In the postprandial period, the rate of glucose clearance depends on the insulin-sensitive tissues. In diabetes, the postprandial state is characterized by a significant and prolonged increase in blood glucose concentration. The postprandial peak blood glucose levels are closely related to atherosclerosis and cardiovascular complications.

A differential diagnosis is made based on the 2-h glucose concentration:

  • Impaired glucose tolerance (IGT), which is an indicator of prediabetes and insulin resistance.
  • Diabetic glucose tolerance, which is usually indicative of the presence of T2D.

The prevalence of IGT is higher in women than in men; with IFG the reverse is the case. The higher prevalence of IFG in men is reported to be due to the fact that men have a lower hepatic insulin sensitivity than women. The higher prevalence of IGT in women is reported to be due to their shorter body height and different fat distribution with the same glucose load /10/. Diagnostic value of the oGTT

Glucose intolerance proceeds continuously within the wide range of normal glucose tolerance up to pathological glucose levels. In this process, the two pathophysiologically important disorders, insulin resistance and β-cell dysfunction progressively worsen. Compared to individuals with normal glucose tolerance, those with IGT in the top quartile of glucose levels only have 20% of their β-cell function left /11/.

Only 28% of 25,000 patients who underwent an oGTT met both IGT and IFG criteria /12/. Lean elderly individuals were more likely to have IFG while obese middle-aged individuals were more likely to have IGT. The prevalence of newly detected T2D cases according to different criteria is shown in Fig. 3.5-1 – Prevalence of newly detected type 2 diabetes, impaired glucose tolerance and impaired fasting glucose and individuals with normal oGTT. Cardiovascular disease and oGTT

One of the main goals in the diagnosis of prediabetes and T2D is the prevention of long-term complications. Epidemiological studies show that IGT is an earlier predictor of future diabetic complications than IFG. This applies to the assessment of the risk of cardiovascular disease (CVD) /13/, increased mortality /14/, and macrosomia /15/. For example, women with isolated postprandial hyperglycemia have a 3.2-fold higher risk for CVD /16/. A high percentage of patients with acute coronary syndrome (ACS) have a pathological oGTT at admission to hospital. A study /17/ showed that 32% of patients with ACS had T2D, 37% had IGT, and 8% had isolated IFG. Hepatic steatosis and oGTT

In obese adolescents with hepatic steatosis, the 2-h oGTT glucose level increases with increasing steatosis. Approximately 73% of those with extended fatty liver met the criteria of metabolic syndrome /18/. A high prevalence of hepatic steatosis in association with T2D was found in adults /19/. Gestational diabetes mellitus (GDM)

One-step strategy: a 75-g oGTT is performed /7/:

Glucose measurement when patient is fasting and at 1 and 2 hours. The diagnosis of GDM is made when any of the following plasma glucose values are met or exceeded:

  • Fasting: 92 mg/dL (5.1 mmol/L)
  • 1 h: 180 mg/dL (10.0 mmol/L)
  • 2 h: 153 mg/dL (8.5 mmol/L) in capillary plasma

and no factors influence the glucose tolerance test (Tab. 3.5-2 – Factors influencing glucose tolerance). The criteria of the oGTT depend on the specimen collection (Tab. 3.5-3 – Criteria of 75 g oGTT depending on specimen selection).

Two-step strategy

A 50-g oGTT with plasma glucose measurement at 1 h, at 24–28 weeks of gestation in women not previously diagnosed with overt diabetes is performed. A 100-g oGTT is performed if plasma glucose level measured 1 h after load are met or exceeded 130 mg/dL, 135 mg/dL, or 140 mg/dL (7.2 mmol/l, 7.5 mmol/L, or 7.8 mmol/L).

Women who screen positive should then undergo the 100-g oGTT on a subsequent day to confirm the diagnosis of GDM. The ADA recommends performing the 100-g oGTT as a confirmatory test for women who screen positive on the 50-g oGTT. GDM is present when any of the following thresholds are exceeded /1/:

  • Fasting ≥ 95 mg/dL (5.3 mmol/L)
  • 1 h ≥ 180 mg/dL (10.0 mmol/L)
  • 2 h ≥ 155 mg/dL (8.6 mmol/L)
  • 3 h ≥ 140 mg/dL (7.8 mmol/L).

There is only weak diagnostic agreement between the results obtained with the 75-g oGTT and the 100-g oGTT. While the 100-g oGTT was diagnostic for GDM in 60 out of 484 pregnant women, the 75-g oGTT only identified 26 women as having GDM.

3.5.5 Comments and problems

Test procedure /20/

The oGTT provides only acceptable reproducibility. The main reason is the large inter individual variability of the glucose concentration, the influence of gastric emptying following ingestion of the hyper osmolar glucose solution, the ambient temperature, and imprecision of the glucose measurement.

The oGTT shows a wide range of variation in the 2-h glucose value in repeat tests, which is usually due to the failure to strictly adhere to the test conditions. Thus, the reproducibility was 49% in pre diabetics, 73% in diabetics, and 93% in normal individuals /21/. Differences in the speed of drinking and incorrect timing of blood collection after 2 h can lead to variations in the blood glucose concentration of up to 20% /22/. This leads to incorrect classification especially in the threshold range. As a result, a pathological result may not always be confirmed by the repeat test.

Factors influencing the oGTT

Refer to:


1. American Diabetes Association. Classification and diagnosis of diabetes: standards of medical care in diabetes –2018. Diabetes Care 2018; 41, Suppl 1: S13-S27.

2. World Health Organisation. Definition, diagnosis and classification of diabetes mellitus and its complications: Report of a WHO consultation. Part 1: Diagnosis and classification of diabetes mellitus. Geneva; World Health Organisation: 1999.

3. Standards of medical care in diabetes – 2012. Diabetes Care 2012; 35, Suppl 1: S11–S63.

4. Executive Summary: Standards of medical care in diabetes: Diabetes Care 2012; 35, Suppl 1: S3–S10.

5. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2012; 33, Suppl 1: S64–S71.

6. American Diabetes Association. Standards of medical care in diabetes. Diabetes Care 2006; 29, Suppl 1: S4–42.

7. Deutsche Diabetes-Gesellschaft. Praxis-Leitlinien. Diabetes and Stoffwechsel 2002; 11 Suppl 2: 1–37.

8. Bartoli E, Castello L, Sainaghi PP, Schianca GPC. Progression from hidden to overt type 2 diabetes mellitus: significance of screening and importance of the laboratory. Clin Lab 2005; 51: 613–24.

9. Ryan J, Siriwardhana D, Vasikaran SD. An audit on oral glucose tolerance test at a large teaching hospital: indications, outcomes, and confounding factors. Ann Clin Biochem 2009; 46: 390–3.

10. Faerch K, Borch-Johnson K, Vaag A, Jorgensen T, Witte DR. Sex differences in glucose levels: a consequence of physiology or methodological convenience? The Inter99 Study. Diabetologia 2010; 53: 858–65.

11. DeFronzo RA, Banerji MA, Bray GA, Buchanan TA, Clement Ss, Henry RR, et al. Determinants of glucose tolerance in impaired glucose tolerance at baseline in the Actos Now for Prevention of Diabetes (ACT NOW) study. Diabetologia 2010; 53: 435–45.

12. DECODE Study Group on behalf of the European Diabetes Epidemiology Study Group. Will new diagnostic criteria for diabetes mellitus change phenotype of patients with diabetes? Reanalysis of European epidemiological data. BMJ 1998; 317: 371–5.

13. Barzilay JI, Spiekerman CF, Wahl PW, Kuller LH, Cushman M, Furberg CD, et al. Cardiovascular disease in older persons with glucose disorders: Comparison of American Diabetes Association criteria for diabetes mellitus with WHO criteria. Lancet 1999; 354: 622–5.

14. DECODE Study Group. Glucose tolerance and mortality: comparison of WHO and American Diabetes Association diagnostic criteria. The DECODE Study Group. European Diabetes Epidemiology Study Group. Diabetes epidemiology: Collaborative analysis of diagnostic criteria in Europe. Lancet 1999; 353: 617–21.

15. Metzger BE. International Association of Diabetes and Pregnancy Study Groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care 2010; 33: 676–82.

16. Barrett-Connor E, Ferrara A. Isolated postchallenge hyperglycemia and the risk of fatal cardiovascular disease in older women and men. The Rancho Bernardo Study. Diabetes Care 1998; 21: 1236–9.

17. Ilany J, Marai I, Cohen O, Matetzky, Gorfine M, Erez I, et al. Glucose homeostasis abnormalities in cardiac intensive care unit patients. Acta Diabetol 2009; 46: 209–16.

18. Cali AMG, De Oliveira AM, Kim H, Chen S, Reyes-Mugica M, Escalera S, et al. Glucose dysregulation and hepatic steatosis in obese adolescents: is there a link? Hepatology 2009; 49: 1896–1903.

19. Kotronen A, Juurinen L, Hakkarainen A, Westerbacka J, Corner A, Bergholm R, et al. Liver fat is increased in type 2 diabetic patients and underestimated by serum alanine aminotransferase compared with equally obese nondiabetic subjects. Diabetes Care 2008; 21: 165–9.

20. Koehler C, Temelkova-Kirktschiev T, Schaper F, Fücker K, Hanefeld M. Prävalenz von neu entdecktem Typ 2 Diabetes, gestörter Glucosetoleranz and gestörter Nüchternglucose in einer Risikopopulation. Dtsch Med Wschr 1999; 124: 1057–1061.

21. Balion CM, Raina PS, Gerstein HC, Santaguida PL, Morrison KM, Booker L, et al. Reproducibility of impaired glucose tolerance (IGT) and impaired fasting glucose (IFG) classification: a systematic review. Clin Chem Lab Med 2007; 45: 1180–5.

22. Harris PE, Kitange HM, Fulcher G, Alberti KGMM. The oral glucose tolerance test: effects of different glucose loads, reproducibility and the timing of blood glucose measurements. Diab Nutr Metab 1991; 4: 293–6.

3.6 Hemoglobin A1c

Lothar Thomas

Hemoglobin A1c (HbA1c) is glycated hemoglobin in which glucose is attached to the N-terminal valine residue of each β-chain of hemoglobin A /1/. Measurement of HbA1c is integral to the diagnosis and management of diabetes because it reflects the average glucose over the preceding 120 days. The HbA1c is now recommended as a standard of care for testing and monitoring diabetes, specifically the type 2 diabetes /2/. Further the mean HbA1c over time provides a reliable measure of chronic glycemia and correlates well with the risk of long term diabetes complications, so that it is currently considered the test of choice for monitoring and chronic management of diabetes /3/.

3.6.1 Indication

The HbA1c is recommended to be performed at least twice a year in diabetes patients with stable blood glucose levels.

3.6.2 Method of determination

Different methods are available for measuring glycohemoglobins. Often these are chromatographic (affinity chromatography, ion exchange high-performance liquid-chromatography) and immunochemical assays.

Cation exchange chromatography

Principle: glycation results in loss of positive charges on the surface of the Hb molecule. On weak cation exchangers and with increasing ion concentrations and/or decreasing pH, glycohemoglobins are eluted before non glycated hemoglobins. Following separation total Hb and HbA1c are measured separately with a spectrophotometer. The proportion of HbA1c is then calculated as the ratio of HbA1c to total Hb. A methodology commonly used is reversed-phase chromatography, in which mixtures of aqueous buffers and organic solvents are used as the mobile phase.


Principle: hemoglobin glycated at the β-N-terminal valine provides a well-defined epitope for antibodies. The determination can be performed by an enzyme immunoassays using monoclonal or polyclonal antibodies which specifically recognize epitopes consisting of the last 4 to 8 amino acids of the glycated N-terminal end of the β-chain of HbA1c.

In a first step, the N-terminal of the glycosylated β-chain is cleaved off by a peptidase. The epitope is bound by specific antibodies, and determined in a latex enhanced immunoassay or solid-phase immunoassay.

The advantage of immunochemical assays is that there is no interference by abnormal hemoglobins or post translationally modified hemoglobins if a specific antibody is selected. The antibody detects only the first four amino acids of the β-chain, the keto amine bond and the glucose. The glycation at the -N-terminal valine can be accurately determined in sickle cell disease, because (β 6Glu-Val) and HbC (β 6Glu-Lys) only occur from position 6 /5/ (Fig. 3.6-2 – Reference material: Endoproteinase Glu-C cleaves the remaining 140 amino acids from HbA0 and HbA1c at the C-terminal side of glutamine). To prevent interference in the determination of HbA1c by HbS and HbC variants, which occur in over 10% of African-Americans and in people of Arabic and Indian descent, some manufacturers use a special technology. The immunoassays, in which the antibodies are directed against a longer peptide, a second peptidase hydrolyzes the HbA1c of the sample to a glycated pentapeptide which competes with the agglutinator (HbA1c-loaded particle) for the anti-HbA1c antibody, thereby reducing the rate of agglutinator. The immunochemical assays also measure the amount of HbA2c, although the concentration is low and usually insignificant. Standardization of HbA1c

The HbA1c test should be performed using a method that is certified by the U.S. National Glycohemoglobin Standardization Program (NGSP) and standardized or traceable to the Diabetes Control and Complications Trial reference assay /6/. A working group of the International Federation of Clinical Chemistry (IFCC) has developed a reference material for the standardization of HbA1c which meets the requirements of the European Union Directive on in-vitro diagnostic medical devices (IVD) and follows the concept of metrological traceability.

The analyte measured is a hemoglobin molecule having a stable adduct of glucose to the N-terminal valine of the hemoglobin β-chain (βN-1-deoxyfructosyl-hemoglobin) /5/ (Fig. 3.6-2 – The endoproteinase Glu-C cleaves the N-terminal 6 amino acids from HbA0 and HbA1c). Pure HbA1c and pure HbA0 are isolated from human blood and mixed in well defined proportions to produce the certified primary reference material (PRMS) used for the reference measurement procedure. The PRMS values are assigned to secondary reference materials (SRMs) and the SRMs are used by the manufacturers to calibrate their instruments.

Result reporting

The results of the HbA1c determination are reported in SI units (mmol/mol) according to IFCC, and related to NGSP units (% and one decimal) using the IFCC-NGSP master equation:

NGSP-HbA1c (%)= 0.0915 (IFCC-HbA1c) + 2,15

IFCC-HbA1c (mmol/mol Hb)= [HbA1c (%)–2.15 ×10.93] Reference to NGSP

Previously, all tests were related to the NGSP3.6.2.1 whose standard is an HbA1c value determined with chromatographic assays in the Diabetes Control and Complication Trial (DCCT). The IFCC Working Group on HbA1c Standardization prepared primary reference materials of pure HbA1c and HbA0 and developed a reference method for HbA1c. Since the NGSP standard contains impurities, the values measured with the IFCC standardization program are 1.5–2% lower. It was agreed in a consensus statement that the values obtained with the IFCC standard program should be traced back to the NGSP, and HbA1c should be reported both in % and in mmol HbA1c per mol Hb.

Reference method /7/

Principle: the determination is performed in three steps: First, the N-terminal end of the hemoglobin β-chain, which is obtained from washed and lysed erythrocytes of the sample, is cleaved off by endoproteinase Gluc-C. High-pressure liquid chromatography, followed by quantification by electrospray ionization mass spectrometry or capillary electrophoresis. Hb A1c is measured as the ratio of glycated to non glycated N-terminal peptide and is reported as a percentage.

In the last step, the glycated (HbA1c peptide) and non glycated (HbA0 peptide) hexapeptides are quantified using mass spectrometry or capillary electrophoresis and UV detection. The percentage of HbA1c is determined based on the ratio of glycated to non glycated N-terminal hexapeptides of the hemoglobin β-chain.

3.6.3 Specimen

EDTA blood: 1 mL (non-fasting patient)

3.6.4 Reference interval

Refer to Ref. /8/ and Tab. 3.6-1 – Standard interpretation of HbA1c.

3.6.5 Clinical significance

The HbA1c value reflects the average blood glucose level of the past 2–3 months. A multinational study /9/ investigated the relationship between HbA1c and the average blood glucose (AG) over 3 months. The best relationship provided the following linear regression between HbA1c and AG:

AG (mg/dL) = 28.7 × HbA1c (%) – 46.7

AG (mmol/L) = 1.59 × HbA1c (%) – 2.59

Although the relationship between HbA1c and average glucose over the preceding 120 days is linear, wide variation was observed between individuals. An HbA1c value of 6.5%, the threshold for diagnosis of diabetes, was associated with an average glucose from 125–175 mg/dL (6.9–9.7 mmol/l) while HbA1c concentrations ranged between 5.5% and 8.0% when average glucose was 150 mg/dL /9/. See Tab. 3.1-10 – Correlation of plasma glucose level with the HbA1c concentration. Assessment of the HbA1c level

Mathematical models and practical experience show that a glycemic change on day 1 of the 120-day life span of the erythrocytes is not fully reflected in a change in HbA1c levels. A significant increase or decrease in mean blood glucose leads to a relatively rapid and significant change in HbA1c. Regardless of the initial HbA1c, it takes 30–35 days to obtain a mean between the initial value and the new final value. Consequently, it only takes 1–2 weeks, not 3–4 months, before a significant change in blood glucose is reflected in a marked change in the HbA1c value. Although the HbA1c level in principle reflects the average blood glucose levels of the past 120 days, hyperglycemic events in days 1–30 contribute approximately 50% to the final result, while events in days 90–120 only contribute about 10% /10/. The level of HbA1c correlates more strongly with fasting glucose (see Section – Postprandial and post absorptive glucose) than with postprandial glucose.

Physiological and pathophysiological conditions can change HbA1c concentration independently of glucose (e.g., race, advanced age, hemoglobin variants, iron deficiency with and without anemia, red blood cell turnover, chronic kidney disease, and cardiovascular events).

  • A reduced life span, as is the case in hemolytic anemias (autoimmune hemolytic anemia, hereditary spherocytosis, sickle-cell anemia, thalassemia), iron deficiency anemia, and blood loss, shortens the time during which Hb is in contact with glucose in the bloodstream, leading to falsely low HbA1c levels.
  • An increase in the life span of the red blood cells (iron deficiency, vitamin B12 deficiency and folic acid deficiency) prolongs the contact between Hb and the glucose in the bloodstream, leading to falsely elevated HbA1c values.

In conditions associated with increased red blood cell turnover, such as sickle cell disease, pregnancy (second and third trimesters) , hemodialysis, recent blood loss or transfusion, or erythropoietin therapy, only plasma glucose criteria should be used to diagnose diabetes /2/.

Due to the intraindividual changes of the Hb levels and the imprecision of the assays, variations between measurements must be at least 0.5% HbA1c before they can be considered to be of clinical relevance. Therefore there should be an interval of 4 to 6 weeks between two HbA1c measurements.

Since the results of the HbA1c measurements are calculated as a ratio of HbA1c to total hemoglobin, they are not influenced by body position or by sampling, whether venous or capillary blood is collected. HbA1c level and diagnosis of diabetes

The ADA has recommended HbA1c with a cutoff ≥ 6.5% for diagnosing diabetes as an alternative to fasting plasma glucose (FPG) ≥ 126 mg/dL (7.0 mmol/L). FPG and oral glucose tolerance tests are recommended for the diagnosis of diabetes only if HbA1c testing is not possible due to patient factors that preclude its interpretation and during pregnancy /23/. The HbA1c cutoff of ≥ 6.5% was associated with 3.8% false negative predictions, while the majority of false negative patients had borderline FPG (7.0–8.0 mmol/L) and HBA1c (6.0%–6.5%), and therefore belonged to at-risk category on the basis of HbA1c alone criteria /11/. See Tab. 3.6-2 – HbA1c and hyperglycemia. Metabolic control of type 2 diabetes

The HbA1c test should be performed for metabolic control at least twice per year. Apart from providing an assessment of the glycemic status of the past 2–3 months and of the risk of future diabetic complications, the HbA1c per se also contributes to improving glycemic control. Thus, it has been shown that patients who know their HbA1c value and know how to interpret it have better glycemic control than patients who don’t /12/. A position paper by the ADA and the European Association for the Study of Diabetes recommends the following glycemic targets /13/:

  • HbA1c of 6.0–6.5% in selected patients (e.g., those with short diabetes history, long life expectancy, no significant cardiovascular disease)
  • HbA1c < 7.0% in most patients to reduce the incidence of microvascular complications
  • HbA1c of 7.5–8.0% in patients with a history of severe hypoglycemia, extensive comorbid conditions, advanced secondary complications, and limited life expectancy.

3.6.6 Comments and problems

Point-of-care HbA1c

The American Diabetes Association (ADA) guidelines include use of Clinical Laboratory Improvement Amendments (CLIA)-waived point-of-care HbA1c testing for diabetes monitoring but not diagnosis /2/.


During storage of the sample over 3–4 days, the decline of erythrocyte metabolism leads to the formation of glutathione adducts of Hb (HbA1d/HbA3), which can interfere with HbA1c especially if chromatographic assays are used. Hemolysates are more unstable as whole blood.


Over 1,200 hemoglobin variants have been identified; the gene β is involved in about 70% of these. While the vast majority are uncommon or rare, certain Hb variants, particularly HbAS, HbAC, HbAD, and HbAE, occur at relatively high frequencies in some populations. One cannot measure HbA1c in individuals who are homozygous for these common variants or who have HbSC disease because they have no HbA /14/.

Disorders of Hb synthesis can be due to:

  • Reduced or absent α- or β-chain production (α- or β-thalassemia), due to homozygous or heterozygous inheritance
  • Changes in the structure of Hb. Hemoglobin anomalies, also named Hb variants, such as HbS, HbC (Africans), HbE (South East Asians), HbD, or Hb defects such as Hb New York are produced. Homozygous patients for these variants cannot be tested for HbA1c. Patients who are heterozygous for these variants can be tested for HbA1c with suitable methods. These include immunological assays, affinity chromatography, and some reversed-phase HPLC assays. However, immunological assays have also been reported to be susceptible to interference from Hb variants such as Hb Okayama (ASβ2; His/Gln); Hb Graz (ASβ2; His/Leu).
  • In newborns, fetal hemoglobin (HbF) accounts for approximately 80% of total Hb, with the percentage falling to below 1% within the first months of life.

HbF interference in chromatographic assays is likely:

  • In infants up to 9 month of age
  • In adults with a rare HbF persistence
  • In compensatory overproduction of HbF in thalassemia.

Drugs, stimulants, chemicals

Stimulants or environmental chemicals can form adducts with Hb which can then interfere with chromatographic assays in particular. This is the case with acetylsalicylic acid (acetylated hemoglobin) and alcohol (acetaldehyde adducts of Hb). Interference should be taken into consideration in the case of individuals who chronically ingest large amounts of these substances /15/.

Renal failure

Urea partly spontaneously decomposes to cyanate and ammonium ions. Cyanate, in the form of isocyanate, forms stable bonds with numerous proteins by carbamylation. In individuals with renal failure and elevated urea levels, the cyanate concentration increases, leading to the presence of carbamylated hemoglobins, which can interfere with HbA1c if chromatographic assays are used /16/. Patients with renal failure, in particular uremic patients, often have impaired erythrocyte kinetics and a reduced erythrocyte life span. This can complicate interpretation of HbA1c results.

Iron deficiency

The presence of iron deficiency with or without anemia leads to an increase in HbA1c values compared to controls, with no concomitant rise in glucose indices /27/. Iron deficiency is associated with shifts in HbA1c distribution from < 5.5 to ≥ 5.5% /28/.

Biological and analytical variation

HbA1c levels increase by 0.1% per decade after 30 years of age /2/. African Americans with diabetes have significantly higher HbA1c concentrations than white patients. The mean between-group difference was estimated to be approximately 0.65% HbA1c. Black patients heterozygous for the common Hb Variant HbS may have lower HbA1c by about 0.35 than those without this the trait /14/.

Biological variability: intraindividual variability 1.7%, inter individual variability 4% /17/.

Analytical goal: see Tab. 3.6-3 – Analytical goals for an HbA1c method with reference to the IFCC reference method.


Falsely low HbA1c values are the result of /18/:

  • Rapidly developing diabetes
  • Reduced erythrocyte life span or hyper regenerative erythropoiesis (hemolytic anemia, recent blood loss, blood transfusion, erythropoietin therapy, malaria, folic acid or vitamin B12 deficiency).
  • Increase in reticulocyte count higher than 3.2% reduce HbA1c values.

Falsely high HbA1c concentrations are the result of:

  • Iron deficiency and iron deficient anemia, but decline in HbA1c under therapy
  • Renal failure
  • Splenectomy
  • Anti-retroviral therapy in HIV patients.

3.6.7 Pathophysiology

The International Union of Pure and Applied Chemistry has recommended to use the term glycohemoglobin for the spontaneously i.e., non-enzymatically occurring glycation of hemoglobin (Hb). All hemoglobins glycated both at the N-terminal end of the β-chain as well as at other free amino groups are referred to as total glycohemoglobin /19/.The total glycohemoglobin is subdivided into sub fractions depending each on the glycation sites and reaction partners. The native (nonglycated) Hb is A0. The sub fractions (HbA1a1 , HbA1a2 , HbA1ab and HbA1ac ) are produced by glycation of the amino group of the N-terminal amino acid valine of the Hb β-chain with different carbohydrates. The sum of these sub fractions are called HbA1.

Proteins are frequently linked to sugar molecules during various enzymatic and nonenzymatic reactions that alter protein function. Glycosylation refers to an enzyme-mediated modification. Glycation refers to a monosaccharide (usually glucose) attaching nonenzymatically. In the case of Hb the glycation occurs by the reaction between the glucose and the N-terminal end of the β-chain, which forms a Schiff base. During the rearrangement, the Schiff base is converted into Amadori products.

Two steps are important (Fig. 3.6-3 – Reaction scheme of the glycation of N-terminal valine of Hb with glucose and subsequent Amadori rearrangement):

  • In the first step glucose in the open chain format binds to the N-terminal to form an aldimine in a reversible reaction
  • In the second step aldimine is gradually converted into the stable keto amine form (Amadori product), namely glycohemoglobin.

The major sites of Hb glycation are β-Val-1, β-Lys-66 and α-Lys 61. The glycation occurs continuously in vivo. However, as the average plasma glucose increases, so does the amount of glycated Hb in the red cells. Glucose passes from plasma through the red cell membrane and binds to Hb, forming an unstable product called aldimine, which then undergoes an Amadori rearrangement to produce a stable keto amine, namely glycohemoglobin). The life of Hb is defined by the erythrocyte survival time which is relatively constant at 10 to 120 days. The degree of glycation , apart from the life time of erythrocytes, depends essentially on the degree as well as the duration of the blood glucose elevation.

The glycation is irreversible and enzyme reactions for the degradation of hemoglobins are not known. The formation of HbA1c within the red cell, therefore, reflects an estimate of the average level of glucose to which the red cell has been exposed. The rate of formation of HbA1c is directly proportional to the glucose level in the blood and represents integrated values for glucose over the preceding 8–12 weeks before blood sampling. The concentration of HbA1c is relatively constant compared to that of glucose and, unlike glucose, is not influenced by food intake or physical activity. The clinical value of HbA1c was evaluated in the Diabetes Control and Complications Trial, which demonstrated a direct relationship between the blood glucose concentration, measured as HbA1c, and microvascular complications of type 1 diabetes (T1D) /20/. Subsequent studies found that there is also a correlation between HbA1c and microvascular complications of type 2 diabetes (T2D) /21/.


1. Sacks DB, Bebu I, Lachin JM. Refining measurement of hemoglobin A1c. Clin Chem 2017; 63: 1433–5.

2. American Diabetes Association. 2. Classification and diagnosis of diabetes: standards of medical care in diabetes – 2018. Diabetes Care 2018; 41, Suppl 1: S13-S27.

3. Sherwani SI, Khan HA, Ekhzaimy A, Masood A, Sakharkar MK. Significance of HbA1c test in diagnosis and prognosis of diabetes patients. Biomarkers Insights 2016; 11: 95- 104.

4. Kerner W, Brückel J. Definition, Klassifikation und Diagnostik des Diabetes mellitus. Diabetologie 2011; 6: S 107–S110.

5. Schleicher E. Neuer Parameter für die Stoffwechseleinstellung: Durchschnittsglucose statt HbA1c? Klinische Chemie Mitteilungen 2009; 40 (3): 63–7.

6. Hanas R, John G. 2010 consensus statement on the worldwide standardization of hemoglobin A1c measurement. Ann Clin Biochem 2010; 47: 290–1.

7. Hoelzel W, Weykamp C, Jeppsson JO, Miedema K, Barr JR, Goodall I, et al. IFCC reference system for measurement of hemoglobin A1c in human blood and the national standard schemes in the United States, Japan, and Sweden: A method-comparison study. Clin Chem 2004; 50: 166–74.

8. The International Expert Committee. International Expert Committee Report on the role of A1c assay in the diagnosis of diabetes. Diabetes Care 2009; 32: 1327–34.

9. Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ. Translating the hemoglobin A1c assay into estimated average glucose values. Diabetes Care 2008; 31: 1473–8.

10. Goldstein DE, Little RR, Lorenz RA, Malone JI, Nathan D, Peterson CM, Sacks DB. Tests of glycemia in diabetes. Diabetes Care 2004; 27: 1761–73.

11. Khan HA, Ola MS, Alhomida AS Sobki SH, Khan SA. Evaluation of HbA1c criteria for diagnosis of diabetes mellitus: a retrospective study of 12785 type 2 Saudi male patients. Endocr Res 2014; 39: 62–6.

12. Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, Nauck M, et al. Management of hyperglycemia in type 2 diabetes: a patient-centered approach. Diabetes Care 2012; 35: 1364–79.

13. Costorino K, Jovanovic L. Pregnancy and diabetes management: advances and controversies. Clin Chem 2011; 57: 221–30.

14. Welsh KJ, Kirkman MS, Sacks DB.Role of glycated proteins in the diagnosis and management of diabetes: research gaps and future directions. Diabetes Care 2016; 39: 1299–1306.

15. Niederau Cm, Coe A, Katayama Y. Interferences of non-glucose adducts on the determination of glycated hemoglobind. Klin Lab 1993; 39: 1015–23.

16. Flückiger R, Harmon W, Meier W, Loo S, Gabbay KH. Hemoglobin carbamylation in uremia. N Engl J Med 1981; 304: 823–7.

17. Braga F, Dolci A, Montagnana M, Pagani F, Paleari R, Guidi GC, et al. Revaluation of biological variation of glycated hemoglobin (HbA1c) using an accurately designed protocol and an assay traceable to the IFCC reference system. Clin Chim Acta 2011; 412: 1412–6.

18. Lapolla A, Mosca A, Fedele D. The general use of glycated haemoglobin for the diagnosis of diabetes and other categories of glucose intolerance: Still a long way to go. Nutrition, Metabolism & Cardiovascular diseases 2011; 21: 467–75.

19. Sacks DB. Translating hemoglobin A1c into average blood glucose: implications for clinical chemistry. Clin Chem 2008; 54: 1756–8.

20. The Diabetes Control and Complications Trial (DCCT). The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993; 329: 977–86.

21. UK Prospective Diabetes Study Group. UKPDS 33: Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes. Lancet 1998; 352: 837–52.

22. American Diabetes Association. Standards of medical care in diabetes – 2014. Diabetes Care 2014; 37, Suppl 1: S14–S80.

23. Selvin E, Steffens MW, Zhu H, Matsushita K, Wagenknecht L, Pankow J, et al. Glycated hemoglobin, diabetes, and cardiovascular risk in nondiabetic adults. N Engl J Med 2010; 362: 800–11.

24. Selvin E, Ning Y, Steffens MW, Bash LD, Klein R, Womg TY, et al. Glycated hemoglobin and the risk of kidney disease and retinopathy in adults with and without diabetes. Diabetes 2011; 60: 298–305.

25. Riddle MC, Ambrosius WT, Brillon DJ, Buse JB, Byington RB, Cohen RM, et al. Epidemiologic relationships between A1c and all-cause mortality during a median 3.4-year follow-up of glycemic treatment in the ACCORD trial. Diabetes Care 2010; 33: 983–90.

26. Zhang X, Gregg EW, Williamson DF, Barker LE, Thomas W, Bullard KM, et al. A1c level and future risk of diabetes: a systematic review. Diabetes Care 2010; 33: 1665–73.

27. English E, Idris I, Smith G, Dhatariya K, Kilpatrick ES, John WG. The effect of anemia and abnormalities of erythrocyte indices on HbA1c analysis: a systematic review. Diabetologia 2015; 58: 1409–28.

28. Kim C, McKeever Bullard K, Herman WH, Beckles GL. Association between iron deficiency and A1c levels among adults without diabetes in the National Health and Nutrition Examination Survey, 1999–2006. Diabetes Care 2010; 33: 780–5.

3.7 Insulin, C-peptide, proinsulin

Lothar Thomas

Pancreatic beta cells are specialized endocrine cells that continuously sense the concentration of blood glucose and other fuels. In response the beta cells secrete insulin to maintain normal fuel homeostasis. The insulin mRNA is translated as the single chain precursor pre proinsulin, and removal of its signal peptide during insertion into the endoplasmic reticulum generates proinsulin. The conversion of proinsulin leads to equimolar production of insulin and C-peptide.

Proinsulin has a direct biologic effect which is one-tenth as much as that of insulin. An increased serum proinsulin-to-insulin ratio is associated with beta-cell dysfunction.

C-peptide is a proinsulin cleavage product and released from the pancreas in amounts equimolar to insulin. C-peptide has no biologic activity on homologous or heterologous tissue and no ability to modify the action of insulin and/or proinsulin.

Insulin is secreted primarily in response to elevated concentrations of blood glucose and increased concentrations of other fuel molecules (e.g., fatty acids and amino acids). Insulin is in charge of facilitating glucose entry into cells.

3.7.1 Indication

Insulin, C-peptide and proinsulin measurements are indicated:

  • Occasionally to estimate the insulin reserve in diabetics
  • More frequently as part of functional tests for the evaluation the cause hypoglycemia
  • In individuals with suspected prediabetes or diabetic metabolism.


  • Diagnosis and differentiation of hypoglycemia
  • In the homeostasis model assessment (HOMA) for quantifying insulin resistance and β-cell function
  • Determination of insulin sensitivity with the hyperinsulinemic clamp.


  • Differentiation of hypoglycemia
  • Estimation of the insulin reserve in diabetics.


  • Diagnosis of insulin resistance
  • Differentiation of hypoglycemia.
  • Suspicion of insulin-producing islet cell tumor (insulinoma)

3.7.2 Functional tests

Functional tests in which insulin, C-peptide and proinsulin levels are measured are listed in Tab. 3.7-1 – Functional tests for the diagnosis and differentiation of hypoglycemia syndrome.

3.7.3 Method of determination


Immunoassay: Most assays use mouse monoclonal anti-insulin antibodies labeled e.g., with acridinium ester, alkaline phosphatase, or ruthenium. An anti-mouse antibody is coupled to a solid phase and binds the insulin-anti insulin antibody complex. After excess mouse monoclonal anti-insulin antibodies have been washed out, the signal emitted by the solid-phase bound insulin-anti-insulin antibody complex is measured. Many commercial assays are specific for insulin and have little cross-reactivity with proinsulin and proinsulin cleavage products /1/.

In the diagnosis of insulinoma, meaningful information can be obtained with assays that have cross-reactivity with des-31,32-split proinsulin and des-64,65 proinsulin, since insulinomas release fragments of insulin and proinsulin.


Immunoassay: these assays use monoclonal anti-C-peptide antibodies labeled e.g., with acridinium ester, alkaline phosphatase, or ruthenium. These antibodies compete with the C-peptide of the sample for binding to a solid-phase bound antibody. After excess anti-C-peptide antibodies have been washed out, the signal emitted by the bound C-peptide antibody complex is measured. The results generated by different commercial assays do not always agree, especially at higher concentrations of C-peptide /2/.


Immunoassay: two-site immunoassays are described, in which one antibody is bound to the solid phase of the micro titer plate and the other is a soluble anti-proinsulin antibody labeled with acridinium ester. The intact proinsulin assay is specific for intact proinsulin while the total proinsulin assay measures all circulating forms of proinsulin /3/.

3.7.4 Specimen


Serum (separate serum from whole blood within 2 h and store at 4 °C for same-day analysis), EDTA plasma is preferable due to higher stability: 1 mL


Serum (immediately separate serum from whole blood and freeze if analysis is not performed on the same day): 1 mL


Serum (separate serum from whole blood within 2 h and store at 4 °C for same-day analysis): 1 mL

EDTA plasma (store at 4 °C if analysis is not performed on the same day): 1 mL

3.7.5 Reference interval

Refer to Tab. 3.7-2 – Insulin, C-peptide and proinsulin reference intervals.

3.7.6 Clinical significance

The secretion of insulin by the β-cells undergoes considerable physiological variations throughout the day which are reflected in fluctuations in insulin levels between 6 and 100 mU/L (42–700 pmol/L).

A pathological β-cell function can be associated with the following conditions and symptoms:

  • Hyperinsulinemia associated with hypoglycemia
  • Hyperinsulinemia associated with normoglycemia (insulin resistance)
  • Hyperinsulinemia with impaired glucose tolerance (prediabetes)
  • Hypoinsulinemia associated with hyperglycemia (overt diabetes).

The assessment of the functional tests is shown in Hyperinsulinemia associated with hypoglycemia

Hypoglycemia can only develop if glucose consumption exceeds glucose production. In individuals with normal carbohydrate metabolism, a decline in plasma glucose levels to below 72 mg/dL (4.0 mmol/L) will cause insulin secretion to decrease and finally stop. However, due to the counter regulatory glucagon response, hypoglycemia only occurs after extreme stress, as a result of fasting in combination with alcohol consumption, and occasionally in pregnancy. In children, hypoglycemia can develop after only 12 h of fasting. Young women are sometimes diagnosed with asymptomatic hypoglycemia after fasting.

Insulinoma was once thought to be the only cause of the conditions of Whipple’s triad in apparently healthy individuals. This is no longer true, since there are other disorders of metabolism and of the endocrine glands that can lead to Whipple’s triad with inadequate insulin secretion /7/ (see Section 3.2 – Hypoglycemia syndromes). The Whipple’s triad is associated with characteristic clinical symptoms, provoked by fasting hypoglycemia with glucose levels of approximately below 50 mg/dL (2.8 mmol/L), which resolve after glucose is administered. Refer to

The main problem with the diagnosis of hypoglycemia lies in its actual definition. It can be defined as glucose levels /8/:

  • In the range of those of fasting healthy individuals
  • In the range of those of patients with insulinoma
  • In a range in which a physiological hormonal response occurs
  • In a range in which clinical autonomic or neuroglycopenic symptoms first develop.

Since the threshold concentration of glucose is the cutoff for the decision as to when there is inadequate insulin response in hypoglycemia, it is important to know the thresholds for adequate hormone secretion.

According to two studies, these are as follows:

  • At a glucose concentration ≥ 40 mg/dL (2.2 mmol/L) in venous plasma after a 72-h fast; below 36 pmol/L for insulin (measured using an unspecific radioimmunoassay), below 200 pmol/L for C-peptide, and below 5 pmol/L for proinsulin /79/
  • At a glucose concentration ≥ 45 mg/dL (2.5 mmol/L) after an 18-h fast and subsequent stationary cycling; below 30 pmol/L for insulin (specific assay), below 100 pmol/L for C-peptide, and below 20 pmol/L for proinsulin /7/.

Functional tests

To associate a hypoglycemia syndrome with hyperinsulinism and to carry out a diagnostic workup, functional tests, such as the fasting test (up to 72 h, or 18 h of fasting combined with cycling), C-peptide suppression test, tolbutamide test, and glucagon test, are performed depending on the patient’s situation. During these tests, it may be necessary to determine insulin, C-peptide, proinsulin, β-hydroxy butyrate and sulfonylurea concentrations. The additional determination of C-peptide levels is performed in order to confirm hyperinsulinism and to establish whether it is due to endogenous or exogenous causes. Hypoinsulinemia associated with hyperglycemia

Based on the presence of autoantibodies (A+) or absence of them (A) and the presence of β-cell functional reserve (β+) or absence of β-cell functional reserve (β), there are four subgroups of diabetes mellitus /16/:

  • A+ and β patients with autoantibodies but no β-cell reserve (insulin reserve)
  • A+ and β+ patients with autoantibodies and preserved β-cell reserve (insulin reserve)
  • A and β patients without autoantibodies and without β-cell reserve (insulin reserve).
  • A and β+ patients without antibodies with β-cell reserve (insulin reserve).

Patients A+β and Aβ are genetically and immunologically different, but have the same clinical picture of type 1 diabetes. Patients A+β+ and Aβ+ are also immunologically and genetically different, but have the same clinical picture of type 2 diabetes with preserved insulin reserve.

The autoantibody diagnosis and determination of β-cell reserve are important classification and prognostic criteria for diagnosing the different phenotypes of diabetes mellitus. At initial manifestation, the presentation is as follows:

  • LADA and type 1.5 diabetes (see Tab. 3.1-1 – Classification of diabetes) as well as the slowly progressing type 1 with autoantibodies but preserved β-cell reserve. LADA excludes patients who become dependent on insulin within the first six months after diagnosis.
  • Phenotype A+β is always insulin dependent, and type A+β+ is insulin dependent in 90% of cases, if ketoacidosis is present as initial manifestation. Determination of β-cell functional reserve

This determination allows clinicians to predict the short-term course of the disease. In the case of initial manifestation of diabetes mellitus and the possible presence of diabetic ketoacidosis (DKA), β-cell functional reserve is determined within 2–10 weeks following the correction of the DKA. For this purpose, baseline C-peptide levels are determined in fasting blood, and the time course of C-peptide levels in the glucagon test. Glucagon test

In this test, blood is collected in the fasting state to determine baseline insulin levels, then 1 mg of glucagon is administered intravenously and blood is collected again at 5 and 10 min. to determine peak levels. The interpretation is shown in Tab. 3.7-6 – Interpretation of baseline and glucagon-stimulated C-peptide levels in diabetics. If insulin reserve (β-cell reserve) is positive, it can be expected to be preserved for 6 months to 1 year, depending on the phenotype. Insulin resistance and β-cell dysfunction

Insulin resistance

Insulin resistance is a condition in which the body produces insulin, but the tissues, in particular muscle, liver and fat tissues, are insensitive to insulin, so that extra insulin is required to transport glucose into the cells. To compensate for the insulin resistance, the islet cells produce extra insulin over many years. Insulin resistance is a criterion of metabolic syndrome that can lead to type 2 diabetes by causing β-cell dysfunction. The following tests are important for early diagnosis of insulin resistance and β-cell dysfunction:

  • Fasting proinsulin levels /17/. Levels above 11 pmol/L are indicative of diminished β-cell function caused by hyperglycemia-induced over stimulation. In this case, the cleavage capacity of carboxypeptidase H and other enzymes is exhausted, causing increasing amounts of unprocessed proinsulin to be released into the circulation /19/.
  • The homeostasis model assessment (HOMA) for estimating insulin resistance (HOMA-IR) and β-cell function (HOMA-β). See Tab. 3.7-7 – Tests for the assessment of β-cell function and insulin resistance.

3.7.7 Comments and problems

Method of determination

Insulin: immunometric assays use monoclonal antibodies and have less cross-reactivity with proinsulin and proinsulin split products than the previously used radioimmunoassay (38–100%). Most assays have less than 2% cross-reactivity with proinsulin, 3% with des-31,32-split proinsulin, but over 40% with des-64,65 proinsulin. This is important in type 2 diabetes as well as impaired glucose tolerance, renal failure and liver cirrhosis, where the concentrations of proinsulin and proinsulin split products are several times higher than in healthy individuals.

However, the disadvantage of using specific assays for the diagnosis of insulinoma is that they do not detect the possibly increased amounts of proinsulin and its split products, making it more difficult to diagnose inadequate insulin secretion /1/. They also may not, or not fully, detect synthetic insulin such as lispro and can miss factitious hypoglycemia /8/.


Insulin: the National Institute of Biological Standards and Controls (NIBSC) offers two standards which are used by the diagnostics manufacturers in the production of kit calibrators:

  • First National Reference Preparation for Insulin 66/304, details of unit amounts are not yet available.
  • First International Standard for Human Insulin 83/500. 1.0 g contains 26,500 units of insulin.
  • Based on amino acid analyses, a conversion factor of 1 mU = 6.0 pmol is recommended. This conversion factor must, however, not be used for the preparation 66/304. In Great Britain, factors in the range of 6.0–7.5 are used, irrespective of the reference preparation /8/.

C-peptide: the first reference preparation has the code 84/510. The new standard preparation 76/561 is a synthetic 64-formyl lysin C-peptide and includes four extra basic amino acids, two at each end. The presence of formyl lysine at position 64 is reported to improve stability and solubility in aqueous solution /8/.

Proinsulin: the NIBSC offers a synthetic proinsulin, code 84/611, which is used for the production of most of the calibrators used in commercial assays /8/. A traceability scheme is proposed /35/.


Insulin: Up to 24 h in EDTA blood at 20 °C, up to 1 week at 4–8 °C, at least 3 months at –20 °C /27/.

C-peptide: C-peptide fragments are produced in vitro which react differently with the antibodies of the various immunoassays. Separation of the serum and storage at –20 °C if the analysis is not performed on the same day is recommended. Even at –20 to –25 °C, levels decrease by 2–26% or 28% within 4 weeks, depending on the assay /8/. Therefore, the sample must always be placed on ice immediately after collection for transport to the laboratory /8/.

Proinsulin: centrifuge within 4 h and store at 4 °C, better –20 °C /28/.


Insulin: hemolysis results in the degradation of insulin due to the release of acid protease (EC /29/.

C-peptide: hemolysis has no influence /29/.

Proinsulin: no effect has been observed. The acid protease has only little influence on proinsulin.


For the assessment of insulin levels, the C-peptide concentration is important. Due to the longer half-life of C-peptide in plasma, insulin fluctuations caused by intermittent secretion can be better evaluated. In addition, the endogenous origin of increased insulin secretion can be confirmed based on the C-peptide concentration /8/.

The US Centers for Medicare and Medicaid Services require a C-peptide test for insulin pump approval. Due to the lack of comparability between the different assays, coverage for insulin pumps is provided only for patients whose C-peptide concentration is below the lower limit of the reference interval of the relevant assay, plus 10% for imprecision of the assay /28/.

The half-lifes and molecular weights are as follows:

  • Insulin 3–4 min., MW 5808 Da
  • Proinsulin approximately 17 min., MW 9390 Da
  • C-peptide 30–40 min., MW 3018 Da.

3.7.8 Pathophysiology 

The secretion of insulin depends on the concentration of blood glucose, gastrointestinal hormones, islet cell hormones, and influence by the autonomic nervous system. Insulin is synthesized by the pancreatic islet cells which consist of the centrally located β- and peripheral α- and δ-cells. The α-cells synthesize glucagon, the β-cells insulin, and the δ-cells insulin growth factor.

In the pancreatic β-cell, the ATP-sensitive K+ channel plays an essential role in coupling membrane excitability with glucose-stimulated insulin secretion. An increase in glucose metabolism leads to elevated intracellular ATP/ADP ratio, closure of K+ ATP channels, and membrane depolarization. Consequent activation of voltage-dependent Ca2+ channels causes a rise in Ca2+ concentration, which stimulates insulin release. Conversely, a decrease in the metabolic signal is predicted to open K+ ATP channels and suppress the electric trigger of insulin secretion. Alterations in the metabolic signal, in the sensitivity of K+ ATP channels, could each disrupt electrical signalling in the β-cell and alter insulin release. Reduced or absent K+ ATP channel activity in the β-cell is causual in congenital hyperinsulism, a rare mostly recessive disorder /30/.

Glucose, hormones and certain drugs that reach the β-cells via arterial blood may cause release of insulin. This in turn is coupled to glucose metabolism. The higher the concentration of glucose, the more it is metabolized by glycolysis in the β-cell. This process produces signals that cause insulin secretion from the secretory granules and synthesis of new insulin /31/.

Like all neuroendocrine peptides, insulin is produced in the endoplasmic reticulum primarily as a single-chain precursor, and than is transported after folding and disulfide bonding to the secretory granules of the Golgi apparatus (Fig. 3.7-5 – Biosynthetic pathway of insulin). The precursor encoded in the Insulin gene is proinsulin, a molecule of 110 amino acids. It consists of a signal peptide that serves to direct the nascent polypeptide chain into the endoplasmic reticulum. There the signal peptide is cleaved off within 1 min., producing proinsulin which folds after approx. 20 min. before it is transported into the Golgi apparatus for storage. In the Golgi apparatus it is packed into the secretory granules together with two proprotein convertases (PC) and carboxypeptidase E (CPE). After approx. 2 h, the secretory granule has its final form and waits for secretion for hours or days /32/.

The granule is secreted in a complex reaction following stimulation of the β-cell by glucose, amino acids, fatty acids, and medications, such as sulfonylureas, which activate voltage-dependent Ca2+ channels by depolarization of the cell membrane potential. In the granules, the PC and CPE are autocatalytically activated. The combined action of the convertases produces des-64,65-split proinsulin, des-31,32-split proinsulin, insulin, and C-peptide (Fig. 3.7-6 – Processing of proinsulin to insulin and C-peptide). Insulin and C-peptide are released in equimolar amounts.

The processing of proinsulin shown in Fig. 3.7-6 does not take place completely. In healthy individuals, about 3% of the proinsulin is not converted to insulin and is released into the bloodstream together with the insulin. The same applies to the proinsulin split products produced in the processing. Elevated concentrations of proinsulin and proinsulin split products are measured in patients with insulinoma, type 2 diabetes, MODY (maturity-onset diabetes of the young), and pancreatic tumors associated with multiple endocrine type 1 neoplasia. Due to the longer half-life of proinsulin (17 min.) compared with insulin (2–3 min.), proinsulin accounts for approximately 10% of immunologically active insulin in plasma of healthy individuals. The long half-life of proinsulin and its low activity (3%) compared to insulin are due to low receptor affinity which is attributable to the connective peptide. The low clearance of proinsulin and its split products compared with insulin leads to their disproportionate increase in relation to secretion. Elevated des-31,32-split proinsulin levels have been found to be of diagnostic importance in the detection of β-cell stress due to over stimulation, such as occurs in type 2 diabetes (e.g., due to elevated concentrations of glucose and fatty acids).

C-peptide is biologically inactive. In contrast to insulin, it is not significantly metabolized by the liver; its main area of distribution is plasma. Although the plasma C-peptide level reflects β-cell function, it is only a limited marker of insulin secretion. The disadvantage is that, due to the (10 times) longer half-life of C-peptide compared to insulin, rapid changes in insulin secretion are not detected in the C-peptide test.

Insulin and C-peptide are secreted into the portal vein, and 40–60% of the secreted insulin is immediately absorbed by the liver. Regular oscillations of insulin secretion occur at 5 to 10 min. intervals. In non diabetics, these are in the range of approximately 3–7 mU/L /33/.

A distinction is made between the following effects of insulin action:

  • Short-term effects occur within minutes at the plasma membrane (e.g., transport of glucose, amino acids, ions) and intracellular catalyzed enzymes. They serve to maintain glucose homeostasis, act directly at the cell membrane, increase the transport of glucose, amino acids and K+ and cause the activation of cytoplasmic enzymes such as of pyruvate dehydrogenase, glycogen synthase, acetyl-CoA-carboxylase, and phosphorylases.
  • Insulin-mediated long-term effects. They require hours to days and include DNA and protein synthesis, the regulation of specific gene expression, and cell growth.

The effects of insulin on skeletal muscle cells, hepatocytes and adipocytes are mediated by receptors of heterogenous structure and function /34/.

The insulin receptor of the cell membrane has a heterotetramer structure consisting of two α-subunits of 135 kDa. These are connected with two transmembrane β-subunits with a MW of 95 kDa via disulfide bridges. If insulin binds to the α-subunit, conformational changes occur which are transferred on to the β-subunit. This leads to activation of the cytoplasmic enzyme tyrosine kinase, which passes on the signal via several postulated mechanisms.

Glucose-stimulated insulin secretion occurs in two phases. During the early phase lasting from seconds to about 10 min., preformed insulin is released from the secretory granules. Then, after a delay lasting from minutes up to 2 h, newly synthesized insulin is secreted, with plasma levels up to > 100 mU/L (700 pmol/L). The reduction in early insulin secretion can be an indicator of the functional disorder of the β-cells and an early symptom of diabetes which may develop in months or years later.

Type 1 diabetes is due to the increasing immune-mediated destruction of β-cells causing a decrease in insulin secretion to less than 1% of normal levels.

Type 2 diabetes is a complex and heterogenous disorder with the following mutually reinforcing or causative and probably consecutive manifestations:

  • Loss of regular oscillations in insulin secretion, which leads to down regulation of the insulin receptors and impaired glucose tolerance and explains why pulsatile secretion of insulin has a greater hypoglycemic effect than continuous secretion /34/
  • Diminished insulin secretion in relation to hyperglycemia after carbohydrate intake
  • Insulin resistance (see Section 3.1 – Diabetes and prediabetes).
  • Increased hepatic glucose synthesis despite the presence of hyperglycemia. This is caused by the stimulation of gluconeogenesis and glycogenolysis due to increased concentrations of free fatty acids.

The WHO hypothesizes that the clustering of type 2 diabetes, hypertension, dyslipidemia, and cardiovascular disease results from the common antecedent – insulin resistance. Insulin resistance across several organs (e.g., adipose tissue, muscle, liver, intestine) results in the metabolic syndrome phenotype /36/.


1. Marcovina S, Bowsher RD, Miller WG, Staten M, Myers G, Caudill SP, et al. Standardization of insulin immunoassays: report of the American Diabetes Association Workgroup. Clin Chem 2007; 53: 711–6.

2. Wiedemeyer HM, Polonsky KS, Myers GL, Little RR, Greenbaum CJ, Goldstein DE, et al. International comparison of C-peptide measurements. Clin Chem 2007; 53: 784–7.

3. Pfützner A, Kunt T, Langenfeld M, Löbig M, Knesovic M, Forst T. Clinical and laboratory evaluation of specific chemiluminescence assays for intact and total proinsulin. Clin Chem Lab Med 2003; 41: 1234–8.

4. Francois A, Maumus S, Vincent-Viry M, Gueguen R, Siest G, Visvikis S. Age and sex-related reference values for serum insulin concentration and its biological determinants in a French healthy population. The Stanislas cohort. Clin Chem Lab Med 2004; 42: 1140–9.

5. Bushhouse SA, Goetz FC, Jacobs DR, Bender AP, French LR, Oestreich PG, Geisser MS. C-peptide response to meal challenge in nondiabetic and diabetic adults living in Wadena, Minnesota. Diabetes Care 1992; 15: 1335–47.

6. Pfützner A, Pfützner AH, Kann PH, Stute R, Löbig M, Yang JW, et al. Clinical and laboratory evaluation of a new specific ELISA for intact proinsulin. Clin Lab 2005; 51: 243–9.

7. Marks V, Teale JD. Hypoglycemic disorders. Clin Lab Med 2001; 21: 79–97.

8. Clark PM. Assays for insulin, proinsulin(s) and C-peptide. Ann Clin Biochem 1999; 36: 541–64.

9. Service FJ. Hypoglycemic disorders. N Engl J Med 1995; 332: 1144–52.

10. Service FJ, O’Brien PC, Kao PC, Young WF Jr. C-peptide suppression test: effect of gender, age and body mass index: implications for the diagnosis of insulinoma. J Clin Endocrinol Metab 1992; 74: 204–10.

11. Boukhman MP, Karam JM, Shavwe J, et al. Localisation of insulinomas. Arch Surg 1999; 134: 818–22.

12. Kao PC, Taylor RL, Service FJ. Proinsulin by immunochemiluminometric assay for the diagnosis of insulinoma. J Clin Endocrinol Metab 1994; 78: 1048–51.

13. Brun JF, Fedou C, Mercier J. Postprandial reactive hypoglycemia. Diabetes and Metabolism (Paris) 2000; 26: 337–51.

14. Hoegsberg B, Gruppuso PA, Couston DR. Hyperinsulinemia in macrosomic infants of nondiabetic mothers. Diabetes Care 1993; 16: 32–6.

15. Hussain K. Congenital hyperinsulinism. Semin Fetal & Neonatal Medicine 2005; 10: 369–76.

16. Greenbaum CJ, Harrison LC, Guidelines for intervention trials in subjects with newly diagnosed type 1 diabetes. Diabetes 2003; 52: 1059–65.

17. Pfützner A, Kunt T, Hohberg C, Mondok A, Pahler S, Konrad T, et al. Fasting intact proinsulin is a highly specific predictor of insulin resistance in type 2 diabetes. Diabetes Care 2004; 27: 682–7.

18. Eisenbarth GS, Conelly J, Soeldner JS. The “natural” history of type 1 diabetes. Diabetes Metabolism Rev 1987; 3: 873–91.

19. Pfützner A, Kann PH, Pfützner AH, Kunt T, Larbig M, Weber MW, Forst T. Intact and total proinsulin: new aspects for diagnosis and treatment of type 2 diabetes mellitus and insulin resistance. Clin Lab 2004; 50: 567–73.

20. De Fronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 1979; 237: E 214–23.

21. Stern SE, Williams K, Ferrannini E, DeFronzo RA, Bogardus C, Stern MP. Identification of individuals with insulin resistance using routine clinical measurements. Diabetes 2005; 54: 333–9.

22. Gungor N, Ardlanian S. Progressive beta cell failure in type 2 diabetes mellitus of youth. J Pediatr 2004; 144: 656–9.

23. Matthews DR, Hosker JP, Rudenski AS, et al. Homeostasis model assessment: insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man. Diabetes 1985; 28: 412–9.

24. Manley S, Luzio SD, Stratton IM, Wallace TM, Clark PMS. Preanalytical, analytical, and computational factors affect homeostasis model assessment estimates. Diabetes Care 2008; 31: 1877–83.

25. Katz A, Nambi S, Mather K, Baron AD, Follmann DA, Sullivan G, et al. Quantitative insulin sensitivity check index: a simple accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab 2000; 85: 2402–10.

26. Bartoli E, Castello L, Sainaghi PP, Schianca GPC. Progression from hidden to overt type 2 diabetes mellitus: significance of screening and importance of the laboratory. Clin Lab 2005; 51: 613–24.

27. Luzio SD, Lucas I, Owens SR. Importance of validation of immunoassays for intact proinsulin. Clin Chem Lab Med 2001; 39: 631–3.

28. Siebenhaar R, Weise A, Safinowski M, Reisinger K, Musholt PB, Reimer T, Pfützner A. Clinical and laboratory evaluation of a new specific ELISA for intact proinsulin. Diabetes, Stoffwechsel, Herz 2008; 4: 275–81.

29. O’Rahilly S, Burnett MA, Smith RF, Darley JH, Turner RC. Haemolysis affects insulin but not C-peptide immunoassay. Diabetologica 1987; 30: 394–6.

30. Kostner JC, Permutt MA, Nichols CG. Diabetes and insulin secretion. The ATP sensitive K+ channel (KATP) connection. Diabetes 2005; 54. 3065–72.

31. Ammon HPT. Wege der Nachrichtenübermittlung bei der Stimulation der Sekretion von Insulin. Diab Stoffw 1995; 4: 311–20.

32. Steiner DF. New aspects of proinsulin physiology and pathophysiology. J Pediatr Endocrinol Metab 2000; 13: 229–39.

33. Bergsten P. Pathophysiology of impaired insulin release. Diabet Metab Res Rev 2000; 16: 179–191.

34. Joost HG. Structural and functional heterogeneity of insulin receptors. Cellular Signaling 1995; 7: 85–91.

35. Little RR, Wielgosz RI, Josephs R, Kinumi T, Takatsu A, Li H, et al. Implementing a reference measurement system for C-peptide: successes and lessons learned. Clin Chem 2017, 63: 1447–56.

36. Higgins V, Adeli K. Pediatric metabolic syndrome: pathophysiology and laboratory assessment.eJIFCC201Vol28No1pp025-042.

Table 3.1-1 Classification of diabetes mellitus /2/


Type 1 diabetes (β-cell destruction, usually leading to absolute insulin deficiency)


Immune mediated


Idiopathic (rare in Europe)


Type 2 diabetes (many range from predominantly insulin resistance with relative insulin deficiency to predominantly secretory defect with insulin resistance)


Gestational diabetes mellitus


Genetic defects of β-cell function

– 1.

Chromosome 12, HNF-1α (MODY3)

– 2.

Chromosome 7, glucokinase (MODY2)

– 3.

Chromosome 20, HNF-4α (MODY1)

– 4.

Chromosome 13, insulin promoter factor-1(MODY4)

– 5.

Chromosome 17, HNF-1β (MODY5)

– 6.

Chromosome 2, Neuro D1 (MODY6)

– 7.

Mitochondrial DNA (matern. inher. diab. and deafness)

– 8.

Other defects


Genetic defects in insulin action (1. Type A insulin resistance, 2. Leprechaunism, 3. Rabson-Mendenhall syndrome, 4. Lipoatrophic diabetes, 5. Others)


Diseases of the exocrine pancreas (1. Pancreatitis, 2. Trauma/pancreatectomy, 3. Neoplasia, 4. Cystic fibrosis, 5. Hemochromatosis, 6. Fibrocalculous pancreatitis, 7. Others)


Endocrinopathies (1. Acromegaly, 2. Cushing’s syndrome, 3. Glucagonoma, 4. Pheochromocytoma, 5. Hyperthyroidism, 6. Somatostatinoma, 7. Aldosteronoma, 8. Others)


Drug or chemical induced (1. Vacor, 2. Pentamidine, 3. Nicotinic acid, 4. Glucocorticoids, 5. Thyroid hormone, 6. Diazoxide, 7. β-adrenergic agonists, 8. Thiazides, 9. Dilantin, 10. γ-Interferon, 11. Others)


Infections (1. Congenital rubella, 2. Cytomegalo virus, 3. Other infections)


Uncommon forms of immune-mediated diabetes (1. Stiff-man syndrome, 2. Anti-insulin receptor antibodies, 3. Others)


Other genetic syndromes sometimes associated with diabetes (1. Down’s syndrome, 2. Klinefelter’s syndrome, 3. Turner’s syndrome, 4. Wolfram’s syndrome, 5. Friedreich’s ataxia, 6. Huntington’s disease, 7. Lawrence-Moon-Biedl syndrome, 8. Myotonic dystrophy, 9. Porphyria, 10. Prader-Willi-Labhart syndrome, 11. Others)


Gestational diabetes mellitus

Table 3.1-2 Diagnosis of prediabetes and diabetes mellitus based on ADA criteria /234/

Clinical and laboratory findings

Prediabetes (also refer to Section 3.1.3)

Categories of increased risk for prediabetes (American Diabetes Association, ADA)

  • Impaired fasting glucose (IFG) of 100–125 mg/dL (5.6–6.9 mmol/L), or
  • impaired glucose tolerance (IGT) based on 75-g oral glucose tolerance test (oGTT) of 140–199 mg/dL(7.8–11.0 mmol/L). The oGTT should be performed in the morning after an overnight fast of at least 8 h, or
  • HbA1c 5.7–6.4% (upper reference limit 5.6%, determined using a method certified by the National Glycohemoglobin Standardization Program and standardized to the Diabetes Control and Complications Trial). The NHANES study 1999–2006 found that a 5.7% cutoff had a diagnostic sensitivity of 39–45% and specificity of 81–91% for the identification of patients with IFG above 100 mg/dL (5.6 mmol/L) or IGT (2-h glucose) above 140 mg/dL (7.8 mmol/L). The HbA1c test does not require a fasting sample. In patients with HbA1c of 5.7–6.4%, the diagnosis of diabetes and its preliminary stages should be made by measuring glucose levels based on conventional criteria /4/.

Note: the overlap of patients who are diagnosed with diabetes in different populations based on glucose and HbA1c is variable and small.

Testing should be considered in all adults who are overweight (BMI ≥ 25 kg/m2) and who have one ore more additional risk factors:

  • Physical inactivity
  • First-degree relative with diabetes
  • High-risk race/ethnicity (e.g., African American, Latino, Native American, Asian American, Pacific Islander)
  • Women who delivered a baby weighing > 4.5 kg or who were diagnosed with gestational diabetes
  • Hypertension (blood pressure ≥ 140/90 mmHg) or on therapy for hypertension
  • HDL cholesterol level < 35 mg/dL (0.90 mmol/L) and/or a triglyceride level > 250 mg/dL (2.82 mmol/L)
  • Women with polycystic ovarian syndrome
  • HbA1c ≥ 5.7%, IGT or IFG on previous testing
  • Other clinical conditions associated with insulin resistance (e.g., obesity, acanthosis nigricans)
  • History of cardiovascular disease.

In the absence of the above criteria, testing for diabetes should begin at age 45 years.

If results are normal, testing should be repeated at least at 3-year intervals, with consideration of more-frequent testing depending on initial results (e.g. those with prediabetes should be tested yearly) and risk status.

Diabetes mellitus

Criteria for testing for diabetes in asymptomatic adult individuals

  • HbA1c ≥ 6,5% (upper reference limit 5.6%, determined using a method certified by the National Glycohemoglobin Standardization Program and standardized to the Diabetes Control and Complications Trial), or
  • fasting plasma glucose (FPG) ≥ 126 mg/dL (7.0 mmol/L) after at least 8 h of fasting, or
  • 2-h plasma glucose ≥ 200 mg/dL (11.1 mmol/L) in the 75-g oGTT, or
  • random glucose level ≥ 200 mg/dL (11.1 mmol/L) or clinical symptoms of hyperglycemia or hyperglycemic crisis.

Notes: the diagnosis of diabetes should be made even if a patient’s findings are contradictory (e.g., HbA1c is ≥ 6.5% on two occasions and fasting glucose is below 126 mg/dL (7.0 mmol/L), or vice versa). If the patient has symptoms of diabetes (weight loss, polyuria, polydipsia), the diagnosis should be made primarily by determining glucose levels. This also applies in the case of conditions that cause false HbA1c results (see Section 3.6 – Hemoglobin A1c/4/.

Testing for type 2 diabetes in asymptomatic children

Note: the incidence of type 2 diabetes in individuals aged 18 years and younger has increased significantly over the past few years.


Overweight (BMI > 85th percentile for age and sex, weight for height > 85th percentile, or weight > 120% of ideal for height).

Plus any two of the following risk factors:

  • Family history of type 2 diabetes in first- or second-degree relative
  • Race/ethnicity (e.g., African American, Latino, Native American, Asian American, Pacific Islander
  • Signs of insulin resistance or conditions associated with insulin resistance (acanthosis nigricans, hypertension, dyslipidemia, polycystic ovary syndrome, low birth weight)
  • Maternal history of diabetes or gestational diabetes mellitus during the child’s gestation

Screening should start at age 10 years or at onset of puberty, if puberty occurs at a younger age. Screening should be repeated every 3 years. The diagnostic criteria are the same as for adult diabetes.

Type 1 diabetic children present with acute symptoms and significantly elevated glucose levels. Islet cell autoantibody testing provides an early indication of the risk of developing type 1 diabetes. For glucose and HbA1c the same criteria as for type 2 apply.

Screening for and diagnosis of gestational diabetes mellitus (GDM) /2/

Pregnant women who meet the following criteria are at high risk of developing diabetes:

  • Heavy obesity
  • History of GDM or birth of a significantly overweight child.
  • Known polycystic ovary syndrome.
  • Family history of type 2 diabetes.

According to the ADA, pregnant women need not be screened for GDM if they belong to the following low-risk group: under 25 years of age, normal weight, no family history of diabetes, no abnormal glucose metabolism, no complications with previous pregnancies, not a member of an ethnic minority with an increased incidence of diabetes.

However, doctors may also test all their pregnant patients for diabetes.

According to ADA the screening for undiagnosed type 2 diabetes should be performed during the first prenatal visit. Screen for GDM at 24–28 weeks of gestation in pregnant women not previously known to have diabetes using one step 2-h 75-g oral glucose tolerance test (oGGT). The oGTT should be performed in the morning after an overnight fast of at least 8 h. The diagnosis of GDM is made when any of the following plasma glucose values are exceeded:

  • Fasting ≥ 92 mg/dL (5.1 mmol/L).
  • 1-h ≥ 180 mg/dL (10.0 mmol/L).
  • 2-h ≥ 153 mg/dL (8.5 mmol/L).

Women with diabetes in the first trimester should receive a diagnosis of overt, not gestational diabetes. Determination of HbA1c will not replace 75-g oGGT for hyperglycemia in pregnant women.

Table 3.1-3 Postabsorptive and postprandial glucose in non diabetics and diabetics /1819/

Clinical and laboratory findings

Nondiabetic:post absorptive state

Approx. 6–12 h after the last food intake, i.e., in the post absorptive state, the transition from the postprandial to the fasting state occurs. In this state, plasma glucose levels remain stable and the entry of glucose into the bloodstream equals glucose uptake by the tissues. The plasma glucose concentration is within the reference interval, and insulin secretion is inhibited. This stimulates the secretion of the counter regulatory hormones (glucagon, catecholamines), which induce the release of glucose from the liver. The glucose generation results from stored glycogen (glycogenolysis) and the synthesis of glucose from e.g., triglycerides (gluconeogenesis). While glycogenolysis dominates during overnight fast, gluconeogenesis dominates during longer fasting periods.

The main gluconeogenic substrates are the so-called three-carbon precursors, including glycerol, lactate, and alanine. Glycerol arises from lipolysis of triglycerides; during gluconeogenesis, two molecules of glycerol are condensed into one molecule of glucose. Free fatty acids after separation from triglycerides are metabolized to ketone bodies. Lactate and alanine are produced by glycolysis in skeletal muscle, the red blood cells and the gastrointestinal tract.

In the post absorptive state, glucose enters and leaves the systemic circulation at a rate of approx. 2 mg (12 μmol) per kg of body weight per minute. Approximately 80% of the glucose is absorbed by the red blood cells, the brain, the liver and the gastrointestinal tract regardless of the insulin level. Skeletal muscle is an insulin-sensitive tissue responsible for the majority of insulin-mediated glucose disposal. It takes up little glucose in the post absorptive state, because plasma insulin levels are low. Like the liver, the heart and the kidneys, skeletal muscle derives its energy from the oxidation of free fatty acids. The relationship between fat and carbohydrate metabolism is reciprocal; the oxidation of free fatty acids inhibits glucose uptake in the insulin-sensitive tissues and stimulates the release of glucose from the liver /20/. Important tissues and organs involved in glucose metabolism are shown in Fig. 3.1-5 – Important organs and tissues involved in glucose metabolism in the postabsorptive and postprandial state.

– postprandial state

The postprandial increase in plasma glucose, which starts 5 to 10 min. after food intake, stimulates the secretion of insulin and inhibits the secretion of glucagon and catecholamines. Glucose enters the circulation from both the liver and the gut. The liver extracts as well as releases glucose. In general the liver and gastrointestinal tissues initially extract 10–25% of ingested glucose, the remainder is extracted by the other tissues and organs. Increases in glucose and insulin and decreases in glucagon, lead to the suppression of 60% of hepatic glucose output. The reduction occurs no later than 60–90 min. after food intake. When the release of glucose from the intestine into the bloodstream decreases, the release of glucose by the liver increases again, thus preventing hypoglycemia.

The postprandial glucose uptake also serves to replenish the glycogen stores in the liver. However, 40–60% of the glycogen is produced from gluconeogenic precursors, and the remainder from the glucose absorbed by the intestine.

The postprandial increase in glucose and insulin suppress lipolysis which dominates in the postprandial energy metabolism. The concentration of free fatty acids in the systemic circulation and hence the fatty acid uptake by skeletal muscle decreases. Muscle, a minor user of glucose in the post absorptive period, becomes a major user postprandially. Approx. 50% of the glucose absorbed is oxidized for energy, approximately 35% is stored as glycogen, and 15% is released into the systemic circulation as lactate and alanine, which are then available for hepatic glycogen synthesis.

Prediabetic, diabetic: – post absorptive state

In both groups the production and disposal of glucose are increased in proportion to plasma glucose levels in the post absorptive state. Because of insulin resistance the absolute rate of glucose disposal only moderately declines, which in turn stimulates insulin secretion, resulting in the following situation:

  • Increased hepatic glucose production from lactate, alanine and glycerol, the initial event in the development of fasting hyperglycemia
  • Decreased utilization of glucose, amino acids and lipids by the liver and peripheral tissues
  • Hyperinsulinemia; the android obesity of many pre diabetics and type 2 diabetics is the result of the deranged carbohydrate metabolism.

The fasting plasma glucose (FPG) in the post absorptive state is closely associated with the insulin secretion in pre diabetics and type 2 diabetics. FPG is therefore normal in individuals who only have insulin resistance and in those with early stages of type 2 diabetes. Elevated FPG levels indicate decreased insulin secretion and increased release of glucose by the liver.

– postprandial state

The ingestion of carbohydrates results in an excessive and prolonged increase in plasma glucose in patients with type 2 diabetes and to less severe hyperglycemia in patients with prediabetes. Compared to non diabetics, impaired suppression of postprandial hepatic glucose release causes postprandial hyperglycemia. In addition, postprandial glycogen synthesis in the liver and skeletal muscle is decreased. The delay in insulin secretion may be particularly important, since the dynamics of insulin availability appear to have greater an effect on carbohydrate metabolism as the absolute amount of insulin released /12/.

Table 3.1-4 Types of MODY and their clinical significance /2627/

Clinical and laboratory findings


The GCK gene encodes the intracellular enzyme glucokinase (GCK), which acts as a glucose sensor of the β-cells of the pancreas. In the β-cells and hepatocytes, GCK catalyzes the first step of glucose metabolism and catalyzes the transfer of a phosphate from ATP to glucose, generating glucose-6-phosphate. This is a rate-limiting step in glucose metabolism, since the GCK activity depends on the ambient glucose concentration. As such, in the β-cells, GCK serves a glucose sensor, facilitating insulin release that is both appropriate and proportional to the blood glucose concentration. Heterozygous loss-of-function mutations in the GCK reduce the rate of phosphorylation and the insulin dose-response curve is shifted to the right. As a result the glycemic threshold for insulin release is regulated at a higher set-point but still remains under tight homeostatic control /27/. Since GCK is also expressed in the hepatocyte, hepatic gluconeogenesis is reduced. There are over 600 known loss-of-function mutations in the 10 exons and the promoter of the pancreatic GCK. Homozygous loss-of-function mutations are rare. Due to the complete loss of GCK activity these patients have insulin-dependent diabetes beginning from the neonatal period.

Clinical findings /27/: GCK patients are detected incidentally at routine screening for an unrelated illness or pregnancy. Since hyperglycemia is mild, micro- and macro vascular complications are rare. Usually one parent has mild fasting hyperglycemia of 100–153 mg/dL (5.5–8.5 mmol/L).

Laboratory findings: persistent fasting glucose level of 100–153 mg/dL (5.5–8.5 mmol/L). HbA1c levels are near normal, values > 7.5% (55 mmoL/moL) would suggest an alternative diagnosis. In the 75-g oGTT, the glucose increment (120-min. glucose minus 0-min. glucose) is less than 54 mg/dL (3.0 mmol/L) in 70% of patients. Persistent fasting C-peptide production (stimulated serum C-peptide > 200 pmoL/L). No pancreatic autoantibodies detectable.

Transcription factor MODY – Generally

Mutations in the genes that encode transcription factors, such as hepatocyte nuclear factor-1 alpha (HNF1A), hepatocyte nuclear factor-4 alpha (HNF4A) and hepatocyte nuclear factor-1 beta (HNF1B), can cause MODY. These transcription factors are components of a network that exists in different tissues. During embryonic development they regulate coordinated gene expression. Mutations in genes of these transcription factors change gene expression for proteins that are involved:

  • In glucose metabolism and glucose transport, including the glucose transporter GLUT-2
  • As enzymes in the mitochondrial glucose metabolism.

As a result there is a progressive decline in β-cell mass in the pancreas which is reported to be caused by reduced proliferation and increased apoptosis.


HNF1A mutations cause autosomal dominant diabetes with long-term diabetic complications. In this type of MODY, the location of the mutation in the HNF1A determines the age of diabetes onset. The penetrance of HNF1A mutations is high, with 63% of carriers developing diabetes by 25 years of age, 79% by 35 years and 96% by 55 years.

Clinical findings: the diabetes typically presents in adolescence, micro- and macro vascular complications occur with the same frequency as in T1D and T2D and are signs of poor glycemic control. Since insulin secretion is often still sufficient in the early stage of the disease, many patients have normal fasting glucose. Mutations in HNF1A alter gene expression for proteins involved in glucose transport including the sodium-glucose cotransporter 2. The patients have glycosuria because of a low renal threshold for glucose /27/.

Laboratory findings: patients often have a normal fasting glucose. In the 75-g oGTT the glucose increment is > 90 mg/dL (5.0 mmol/L). Glycosuria at blood glucose > 180 mg/dL (10 mmol/L). Persistent fasting C-peptide production (stimulated serum C-peptide > 200 pmol/L). No pancreatic autoantibodies detectable. HDL cholesterol > 50 mg/dL (1.3 mmol/L).


HNF4A-MODY accounts for 3–5% of MODY cases and, like HNF1A-MODY, presents as a progressive decline in β-cell function. The penetrance of HNF4A mutations is variable, and the majority of mutation carriers develop diabetes by the age of 25 years.

Clinical findings: the HNF4A-MODY presents in the same way to HNF1A-MODY. The identification of an HNF4A mutation has important implications for the management of pregnancy. Mutations in HNF4A are associated with an 800 g increase in birth weight compared to unaffected siblings /27/. Approx. 10% of heterozygous HNF4A mutation carriers have diazoxide-responsive hyperinsulinemic hypoglycemia during the first weeks of life.

Laboratory findings: in the 75-g oGTT the glucose increment is > 90 mg/dL (5.0 mmol/L). Persistent fasting C-peptide production (stimulated serum C-peptide > 200 pmol/L). No pancreatic autoantibodies detectable. Low HDL cholesterol, elevated LDL cholesterol, low triglycerides.


Mutations in the gene encoding the transcription factor HNF1B accounts for 1% of MODY cases. The transcription factor is expressed in the pancreas, kidneys, liver, gut and genital tract. In half of the HNF1B mutation carriers is the result a β-cell dysfunction combined with insulin resistance. Penetrance is variable, diabetes develops between the ages of 1 and 61. A third of all HNF1B mutations are deletions of the whole gene.

Clinical findings: often low birth weight and transient neonatal diabetes due to reduced insulin secretion. The most common extra-pancreatic feature are renal cysts which lead to end-stage renal disease in 50% of cases under the age of 45.

Laboratory findings: in the 75-g oGTT the glucose increment is > 90 mg/dL (5.0 mmol/L). Persistent fasting C-peptide production (stimulated serum C-peptide > 200 pmol/L). No pancreatic autoantibodies detectable. Elevated creatinine, hyperuricemia, elevated aminotransferases, hypomagnesemia /27/.

Other rare types of MODY

Very rare types of MODY which cause autosomal dominant diabetes involve the IPF1, NEUROD1, CEL, INS, KCNJ11, and ABCC8 genes /27/.

Table 3.1-5 Laboratory findings in other types of diabetes

Clinical and laboratory findings

Pancreatic disease – Generally

Depending on the extent of hyperglycemia, long-term diabetic complications such as neuropathy, retinopathy and nephropathy may occur as a result of pancreatic diseases.

– Acute pancreatitis

Inflammatory processes in the pancreatic tissue often cause transient increases in blood glucose levels.

– Chronic pancreatitis

Over 50% of patients with chronic pancreatitis have overt diabetes or impaired glucose tolerance. In a study /32/ of patients with chronic pancreatitis, 35% had insulin-dependent diabetes, 31% were not insulin-dependent but had impaired glucose tolerance, and 34% had normal glucose tolerance. Diabetes was diagnosed if two consecutive fasting capillary blood glucose concentrations > 130 mg/dL (7.2 mmol/L) were measured performed 1 year following the diagnosis of chronic pancreatitis.

– Pancreatectomy

Pancreatectomy results in absolute insulin deficiency as well as glucagon deficiency. The latter causes an increase in the concentrations of free amino acids and fatty acids in plasma. Diabetes resulting from pancreatectomy is associated with low insulin requirement and severely fluctuating blood glucose levels, in particular hypoglycemias.

Liver disease – Generally

Approx. 40–70% of fatty liver patients, 40–60% of those with chronic hepatitis and 60–80% of those with cirrhosis of the liver have impaired glucose tolerance. In patients with chronic liver disease who have fasting glucose levels of 110–140 mg/dL (6.1–7.8 mmol/L) and no clinical symptoms of diabetes, glucose tolerance should be determined using the oGTT.

– Cirrhosis of the liver

Patients with cirrhosis of the liver are 3 to 4 times more likely to develop diabetes than those with a healthy liver. Approximately 60–80% of patients with liver cirrhosis have impaired glucose tolerance, 40% have overt diabetes. The latter has characteristics similar to those of T2D, namely insulin resistance and reduced insulin secretion /33/. Cirrhosis patients mainly have muscle cell insulin resistance, regardless of whether they are glucose tolerant, glucose intolerant or diabetic. The glucose intolerance in patients with liver cirrhosis is more pronounced in the oGTT than in the intravenous GTT.

Hereditary hemochromatosis

Hereditary hemochromatosis is an autosomal recessive disorder of iron metabolism. It occurs with a prevalence of about 4.5 per 1,000, with men being affected five times more frequently than women. The diabetes associated with hemochromatosis corresponds to T2D with a reduced insulin response and normal or increased glucagon response to glucose load. Approximately 20–80% of hemochromatosis patients have diabetes, depending on a study /31/.

Hereditary hemochromatosis occurs more frequently in diabetics /34/. The prevalence is 6.1 per 1,000 in diabetics based on a hepatic iron index of > 2.0. This index is determined by dividing the hepatic iron concentration in μmol/g of dry weight by the patient’s age in years. Screening for hemochromatosis was based on a transferrin saturation of > 55%.

Endocrinological disease – Generally

Hormones of the anterior pituitary (ACTH, hGH), thyroid gland and the adrenal cortex (glucocorticoids) have a diabetogenic effect. Hyper function of these organs can therefore lead to impaired glucose tolerance and diabetes.

– Acromegaly

Acromegaly is a rare disease (annual incidence of 5 per 1 million) caused by excess secretion of human growth hormone (hGH), usually due to a pituitary adenoma. hGH is a powerful anabolic hormone which increases protein synthesis. Its effect is mediated by the insulin-like growth factors (IGFs), also referred to as somatomedins. The elevated hGH levels lead to cellular insulin resistance, which causes impaired glucose tolerance in approximately 15–30% and overt diabetes in approximately the same percentage of acromegaly patients. Acromegaly is confirmed if hGH levels do not fall below 1 μg/L in the oGTT /35/.

– Cushing’s syndrome

Cushing’s syndrome can be caused by excess secretion of ACTH either by a pituitary adenoma, an autonomous adenoma of the adrenal cortex, or due to para neoplastic ACTH production.

Half of all patients with Cushing’s syndrome have diabetes, 90% have impaired glucose tolerance /36/. Fasting blood glucose levels are often normal. The diabetogenic effect of the hyper secreted glucocorticoids is due to inhibited insulin-stimulated uptake of glucose by the tissues.

– Pheochromocytoma

Approximately two thirds of pheochromocytoma patients have impaired glucose tolerance, since catecholamines promote hepatic glycogenolysis and inhibit insulin secretion /37/.

– Conn’s syndrome

Some patients with primary hyperaldosteronism have impaired glucose tolerance due to reduced insulin secretion as a result of total body potassium depletion and due to diminished tissue insulin sensitivity as a result of impaired Na+/K+-ATPase activity /37/.

– Hyperthyreosis

Approximately half of all patients with hyperthyreosis have impaired glucose tolerance. Overt diabetes occurs rarely /52/.

Table 3.1-6 Testing for diabetes in pregnant women /2/

Clinical and laboratory findings

Screening for undiagnosed type 2 diabetes at the first prenatal visit in those with risk factors

The Consensus Panel of the International Association of Diabetes and Pregnancy Study Groups /38/ recommends that pregnant women who are at increased risk of developing diabetes due to factors such as obesity (body mass index ≥ 27 kg/m2, first-degree relatives with diabetes, history of gestational diabetes mellitus (GDM), multiple miscarriages, stillbirth, congenital deformity) be tested for diabetes mellitus as early as during the first prenatal visit /38/. According to the ADA /42/, pregnant women need not be screened for GDM if they belong to the following low-risk group: under 25 years of age, normal weight, no family history of diabetes, no abnormal glucose metabolism, no complications with previous pregnancies, not a member of an ethnic minority with an increased incidence of diabetes. However, doctors may also test all their pregnant patients for diabetes /39/.

During the first prenatal visit, all pregnant women at increased risk should be tested for fasting plasma glucose (FPG) or HbA1c, or should undergo the oGTT. GDM is present if one of the tested biomarkers exceeds the following cutoffs /1/:

  • Fasting plasma glucose ≥ 92 mg/dL (5.1 mmol/L)
  • HbA1c ≥ 6.0% (the level should be below 6.0% even prior to conception).

75-g oral glucose tolerance test:

  • After 1 h ≥ 180 mg/dL (10.0 mmol/L).
  • After 2 h ≥ 153 mg/dL (8.5 mmol/L).

If only fasting glucose testing is used and the level is below 92 mg/dL (5.1 mmol/L), an oGTT is nevertheless performed at 24–28 weeks of pregnancy.

Screen for gestational diabetes (GD) at 24–28 weeks of gestation in pregnant women not previously known to have diabetes

A 75-g oGTT should be performed in the morning after an overnight fast. The diagnosis of GDM is made when any of the following plasma glucose levels are exceeded:

  • Fasting: ≥ 92 mg/dL (5.1 mmol/L)
  • 1-h: ≥ 180 mg/dL (10.0 mmol/L)
  • 2-h: ≥ 153 mg/dL (8.5 mmol/L).

The investigations of the HAPO study (cohort of 25,000 pregnant women) /44/ have shown that in 8.3% GDM was identified based on fasting glucose alone, an additional 5.7% based on the 1-h oGTT value, and a further 2.1% based on the 2-h value. 11.1% of the pregnant women had only one elevated value, 3.9% had two elevated values, and 1.1% had elevation of all three values. The total incidence of GDM was 17.8%. The main diagnostic criteria were fasting glucose and the 1-h oGTT value. The study showed a positive correlation between the fasting glucose level, the 1-h value and the 2-h value and a birth weight above the 90th percentile, the cord C-peptide level, and the infant body fat.

Monitoring GDM – Generally

Glycemic goals in pregnant women in plasma /3/:

  • Preprandial: ≤ 95 mg/dL (5.3 mmol/L)
  • 1-h postprandial: ≤ 140 mg/dL (7.8 mmol/L)
  • 2-h postprandial: ≤ 120 mg/dL (6.7 mmol/L)
  • Pre meal, bedtime and overnight: 60–99 mg/dL (3.3–5.4 mmol/L)
  • Peak postprandial 100–129 mg/dL (5.4–7.4 mmol/L)

If higher levels are measured, insulin therapy should be considered /45/.

HBA1C < 6%

– Screening of women with GDM for persistent diabetes at 6–12 weeks postpartum

In women with GDM, a normal daytime glucose profile on day 3–4 rules out persistent diabetes. In the case of mothers with macrosomic children, the daytime glucose profile will have no diagnostic value until 6 weeks post partum. The pre prandial plasma glucose should be less than 110 mg/dL (6.1 mmol/L) and the 2-h postprandial glucose level less than 140 mg/dL (7.8 mmol/L). Pregnant women with a history of GDM in previous pregnancies should undergo a 75-g oGTT 6 weeks post partum. Levels between 140 and 200 mg/dL (7.8–11.1 mmol/L) are indicative of impaired glucose tolerance, and a level of more than 200 mg/dL (11.1 mmol/L) is indicative of overt diabetes.

– Glycemic control of gestational diabetes

Conventional capillary glucose monitoring

The therapeutic goal is normoglycemia with blood glucose levels as follows /45/:

  • Pre prandial: ≤ 90 mg/dL (5.0 mmol/L)
  • 1-h postprandial: ≤ 130 mg/dL (7.2 mmol/L)
  • 2-h postprandial: ≤ 120 mg/dL (6.7 mmol/L)
  • Pre meal, bedtime, and overnight: 66–99 mg/dL (3.3–5.4 mmol/L)
  • Peak postprandial: 100–129 mg/dL (5.4–7.1 mmol/L) /3/.

The incidence of macrosomia increases significantly if the mean daytime glucose in venous plasma exceeds 120 mg/dL (6.6 mmol/L), corresponding to 140 mg/dL (7.8 mmol/L) in hemolyzed capillary whole blood. During the first half of pregnancy, glucose levels can fluctuate greatly, and hypoglycemia is common. The mean glucose threshold for spontaneous miscarriage is 150 mg/dL (8.3 mmol/L).

In pregnant women with diabetes, the HbA1c value should be less than 7% prior to conception, and less than 6% under therapy /2/. Maternal glycemia during the first 6 to 8 weeks of gestation and elevation in HbA1c of 1% increase the risk of malformations compared to nondiabetic gravidae /2/.

Continuous glucose monitoring (CGM) /84/

In a multi centre, international, randomized controlled trial (CONCEPT) the effectiveness of continuous glucose monitoring (CGM) on pregnant women with type 1 diabetes was compared with conventional capillary glucose monitoring. Findings:

  • Women randomized to CGM had a significant greater time in reduction of HbA1c (68% versus 61%); HbA1c ≤ 6.5%.
  • Women randomized to CGM had reduced hyperglycemia (100 min versus 60 min daily)
  • Women randomized to CGM had reductions in the proportion of infants large for gestational age (53% versus 69%)
  • In women randomized to CGM admission to neonatal intensive care was reduced (27% versus 43%)
  • In women randomized to CGM neonatal hypoglycemia was reduced (15% versus 28%).

Table 3.1-7 Tests for the of risk assessment of infants born to diabetic mothers

Clinical and laboratory findings

Blood glucose

In healthy pregnant women, the glucose level of the fetus and newborn is 10–20% lower than that of the mother /48/.

Newborns of mothers with GDM and manifest diabetes are at high risk of hypoglycemia, since the acquired hyperinsulinism is still present and hepatic gluconeogenesis is reduced. The incidence and extent of hypoglycemia correlate with the quality of metabolic control during the last weeks before delivery. Glucose levels in newborns of diabetic mothers should always be determined between the 1st and 3rd hour of life when glucose levels are expected to be at their nadir. During the first 24 hours of life, plasma glucose levels ≤ 45 mg/dL (2.5 mmol/L) indicate hypoglycemia. Glucose concentrations should be measured in venous plasma. Due to the high hematocrit values in newborns, the results for venous and capillary whole blood glucose are falsely low. The postnatal development of infants born to mothers with GDM shows an increased incidence of impaired glucose tolerance, regardless of the glycemic control of the mother during pregnancy. In children < 5 years, between 5–9 years and 10–16 years of age the incidence of impaired glucose tolerance was 1.2%, 5.4% and 19.6% respectively. The cause is reported to be due to altered fetal β-cell function /49/.

Hematocrit in venous blood or umbilical cord blood

Erythropoietin stimulates the proliferation and differentiation of erythropoiesis in the liver of the fetus in the bone marrow thereafter. An O2 sensor in the fetal liver or in the kidney post partum regulates the synthesis of erythropoietin in such a way that the tissues are sufficiently supplied with O2 /5051/. In mothers with poorly controlled GDM, T1D and T2D, the increased deposition of glycogen in the placenta causes insufficient supply of the fetus with O2. The fetus compensates with increased erythropoietin synthesis, leading to polycythemia.

Neonatal polycythemia is diagnosed based on the hematocrit. Due to the fact that physiologically a rise in hematocrit occurs as early as in the first 6 hours of life, the threshold above which neonatal polycythemia is diagnosed should be dynamic, i.e., related to the time of sampling. A study /59/ defined polycythemia as a 2-h hematocrit > 71%, a 6-h hematocrit > 68%, and a cord blood hematocrit > 56% (sampled within 30 seconds post partum). Approximately 5% of newborns of mothers with GDM or T1D or T2D have polycythemia.


Newborns of mothers with GDM or overt T1D or T2D have a higher incidence of severe neonatal jaundice than newborns of healthy mothers. The causes are polycythemia, macrosomia, and premature birth. Pathological conditions are indicated by bilirubin concentrations rising faster than 5 mg/dl (85 μmol/L)/day and absolute values > 15 mg/dL (257 μmol/L). Further information can be found in Section 5.2 – Bilirubin.

Calcium, magnesium

Newborns of mothers with poorly controlled T1D, T2D or GDM can have hypocalcemia and hypomagnesemia. There are often no clinical manifestations. Occasionally, motor restlessness or even seizures may occur. Hypocalcemia/hypomagnesemia are defined as serum calcium/magnesium concentrations of less than the following concentrations:

  • Total serum calcium: 7.1 mg/dL (1.77 mmol/L)
  • Ionized serum calcium: 4.6 mg/dL (1.17 mmol/ L)
  • Magnesium: 1.2 mg/dL (0.48 mmol/L)

Table 3.1-8 Testing for diabetes and glucose monitoring

Clinical and laboratory findings

Glucose in blood

Test strategies for the diagnosis of diabetes mellitus

According to the criteria of the American Diabetes Association (ADA), diabetes is present when the venous plasma glucose meets one of the following criteria /2/:

a) Clinical symptoms of diabetes and a random plasma glucose concentration ≥ 200 mg/dL (11.1 mmol/L). The relevant thresholds for venous whole blood (hemolyzed), capillary whole blood (hemolyzed) and capillary plasma are 180 mg/dL (10.0 mmol/L), 200 mg/dL (11.1 mmol/L), and 220 mg/dL (12.2 mmol/L) respectively /55/.

b) Fasting glucose in venous plasma and capillary whole blood (hemolyzed) ≥ 126 mg/dL (7.0 mmol/L). No corresponding values have been defined for venous whole blood (hemolyzed) and capillary plasma /55/.

c) A 2-h oGTT glucose concentration ≥ 200 mg/dL (11.1 mmol/l) for capillary whole blood (hemolyzed) and ≥ 200 mg/dL (11.1 mmol/L) for venous plasma. The corresponding thresholds for venous whole blood (hemolyzed) and capillary plasma have been determined as 180 mg/dL (10.0 mmol/L) respectively 220 mg/dL (12.2 mmol/L) and are not officially defined /55/.

d) Clinical symptoms of diabetes and an HbA1c level ≥ 6,5%.

Note: a positive (pathological) result in a) through d) should be confirmed by a repeat test on a subsequent day. This is not required if hyperglycemia is evident or if there are other relevant metabolic symptoms.

Reason for the definition of the thresholds: the 2-h oGTT threshold is the level above which the prevalence of microvascular complications in diabetes (retinopathy and nephropathy) increases significantly. According to a report by the ADA, studies involving Pima Indians, Egyptians, and data from the Third National Health and Nutrition Examination Survey (NHANES III) showed that a fasting plasma glucose level ≥ 126 mg/dL (7.0 mmol/L) is virtually equivalent to a 2-h oGTT level of ≥ 200 mg/dL (11.1 mmol/L) /1/.

Test strategies for the diagnosis of prediabetes

The ADA defines impaired fasting glucose (IFG), impaired glucose tolerance (IGT) and borderline HbA1c as intermediate stages between normal glucose homeostasis and T2D. They are risk factors for T2D and cardiovascular disease. The ADA recommends HbA1c, fasting glucose or oGTT screening in asymptomatic individuals > 45 years of age with a body mass index > 25 kg/m2 and, if test results are normal, repeat testing at 3 year intervals. Younger individuals should be screened if they have risk factors for diabetes, e.g., obesity, family history of diabetes in first-degree relatives, member of an ethnic population with a high incidence of diabetes, birth of an overweight child, blood pressure > 140/90 mmHg, HDL cholesterol < 35 mg/dL (0,90 mmol/l), and triglycerides > 250 mg/dL (2.82 mmol/L).

In addition, screening at 2-year intervals is recommended for overweight children aged 10 years and over /46/. Prediabetes is confirmed if any of the following are present:

a) IFG, i.e., the fasting venous plasma glucose concentration is ≥ 100 mg/dL (5.6 mmol/L) but < 126 mg/dL (7.0 mmol/L). A fasting glucose level below 100 mg/dL (5.6 mmol/L) is considered normal and rules out impaired glucose tolerance /1/. One problem with IFG is that it does not capture the same group of individuals as IGT. Only 28% of individuals meet both IFG and IGT criteria. The differences are particularly significant in the case of lean, older individuals who have normal fasting glucose, but a positive (pathological) 2-h oGTT result. In contrast, middle-aged overweight individuals have elevated fasting glucose levels, but a normal 2-h oGTT result.

b) IGT, i.e., the 2-h glucose concentration from the oGTT is ≥ 140 mg/dL and < 200 mg/dL (7.8 mmol/L and 11.1 mmol/L), measured in venous plasma. The corresponding thresholds for venous whole blood (hemolyzed) are ≥ 120 and < 180 mg/dL (6.7 mmol/L and 10.0 mmol/L), the thresholds for capillary whole blood (hemolyzed) are ≥ 140 mg/dL and < 200 mg/dL (7.8 mmol/L and 11.1 mmol/L), and for capillary plasma they are ≥ 160 mg/dL and < 220 mg/dL (8.9 mmol/L and 12.2 mmol/L) /55/.

c) An HbA1c level between 5.7% and 6.4% indicates a risk of diabetes.

Gestational diabetes (GDM): GDM is defined as impaired glucose tolerance during pregnancy and is diagnosed using the oGTT. See Tab. 3.1-6 – Testing for diabetes in pregnant women and Tab. 3.1-2 – Diagnosis of prediabetes and diabetes mellitus based on ADA criteria.

Oral glucose tolerance test (oGTT)

The oGTT and the fasting glucose are suitable for diagnosing prediabetes and T2D. However, the ADA prefers the HbA1c and fasting glucose, since they are easier to carry out, more acceptable to patients, less costly and less dependent on influencing variables and interfering factors. A glucose level above 200 mg/dL (11.1 mmol/L) 2 hours after a 75-g oGTT indicates diabetes, which is confirmed by a repeat oGTT on another day. A 2-h glucose level ≥ 140 mg/dL (7.8 mmol/L) but < 200 mg/dL (11.1 mmol/L) is defined as IGT (prediabetes), which is a risk factor for T2D. Normoglycemia is present if the 2-h value is < 140 mg/dL (7.8 mmol/L).

The WHO’s indication for oGTT differs from that of the ADA. The WHO recommends performing the oGTT if random glucose is in the range > 110 to 200 mg/dL (6.1–11.1 mmol/L) or fasting plasma glucose is > 110 mg/dL (6.1 mmol/L) and < 126 mg/dL (7.0 mmol/L). In the latter case no lower reference limit is specified below which a diagnosis of diabetes can be safely ruled out.

Urinary glucose

The semi quantitative determination of urinary glucose, once commonly used in population screening for diabetes, is no longer recommended for screening for hyperglycemia or monitoring of diabetes. Even if glucosuria can be detected in patients with severely elevated glucose levels, the determination of urinary glucose does not provide any information regarding the presence of hyperglycemia if its extent is below the renal glucose threshold (180 mg/dL; 10.0 mmol/L). In addition, the glucose concentration in urine highly depends on the volume of urine /46/.

Ketone bodies

The ketone bodies acetoacetate, β-hydroxy butyrate and acetone are produced as by-products in the metabolic breakdown of free fatty acids. They are determined in blood and urine, mainly in the treatment of T1D. In acute cases they can be used as an additional parameter for the diagnosis and monitoring of diabetes. Elevated ketone levels in patients with diabetes or hyperglycemia are diagnostic of diabetic ketoacidosis (DKA). DKA is a medical emergency /45/. Also refer to Section 5.5 – Ketone bodies.


Diagnosis of diabetes mellitus /2/

A level ≥ 6.5% (48 mmol/mol), upper reference limit 5.6% (38 mmol/mol), determined using a method certified by the National Glycohemoglobin Standardization Program and standardized according to the Diabetes Control and Complications Trial. The diagnosis of diabetes in patients with diabetes symptoms should be made primarily on the basis of glucose levels.

Monitoring of diabetes /2/

Non-pregnant diabetic < 7%, pregnant diabetic ≤ 6%; preschool-aged children with T1D < 8.5% (but > 7.5%), children aged 6–12 < 8%, adolescents < 7.5%.

Genetic markers

One of the key focuses in predicting the genetic risk of type 1A diabetes is genotyping of the HLA–DR and HLA–DQ loci. For example, children who carry both high-risk haplotypes (DR3–DQ2 and DR4–DQ8) have a 1 : 20 risk of developing type 1A by the age of 15. Furthermore, a number of non-HLA-associated gene loci that are associated with type 1A were identified. For example, the odds ratio of the INS and PTPN22 loci is approximately 2–2.5. INS is the gene for encoding insulin and the primary auto antigen in type 1A. PTPN22 encodes the tyrosine phosphatase of T-cells and is involved in T-cell receptor signaling /16/. Also refer to Section 3.1.3 – Categories with increased risk for diabetes (prediabetes).

Islet cell antibodies

Approximately 85–90% of T1D patients have islet cell antibodies when fasting hyperglycemia is initially detected. Islet cell antibodies may be present for months to years before the appearance of clinical symptoms. The risk of developing T1D in relatives of patients with the disease is approximately 5%, which is 15-fold higher than the risk in the general population (1 in 250 to 300 lifetime risk). Screening relatives can identify individuals at increased risk for T1D. However, as many as 1–2% of healthy individuals also have at least one detectable islet cell antibody. Because of the low prevalence of T1D (0.3% in the general population), the positive predictive value of a single islet cell antibody will be low. However, the presence of several islet cell antibodies, such as ICA, GADA, IA-2A, and IAA (see Section 25.10 – Autoimmune markers in diabetes mellitus), is associated with a > 90% risk of T1D. Testing for islet cell antibodies can be useful in the following situations:

  • Screening of children with a family history of T1D
  • Differentiation of T2D from T1D in children
  • Testing of adults in whom LADA is suspected
  • Screening of nondiabetic family members who wish to donate a kidney or part of their pancreas for transplantation
  • Decision whether the patient can be treated with insulin or oral anti diabetics.

Type 1B diabetes mellitus: these patients have no islet cell antibodies.

LADA: this type of diabetes is also referred to as latent autoimmune diabetes of adulthood. Up to 10% of Caucasians who clinically present with a T2D phenotype have islet cell antibodies, particularly anti-GAD65, which predict insulin dependency. Although autoantibody-positive patients with T2D progress faster to absolute insulinopenia than do antibody-negative patients, the latter may also become insulin dependent with time.


The determination of albumin in urine is indicated for early detection of diabetic nephropathy, because diabetes is the main cause of chronic kidney failure in Europe and North America /2/. The commonly used reagent strip tests for detecting proteinuria can only detect macro albuminuria, i.e., excretions ≥ 300 mg/L. However, diabetic nephropathy is present in the case of micro albuminuria, which is defined as excretions in the ranges of 30 and 299 mg/24 h, 20–200 μg/min. or 30–300 mg/g creatinine, if two of three urine samples test positive /2/. Excretion rates above these thresholds are termed overt or clinical nephropathy.

Diagnosis: diabetics who tested dip-stick negative for proteinuria should always be tested for micro albuminuria.

  • Children with T1D should be tested for albuminuria 5 years following diagnosis or annually from the beginning of puberty, since albuminuria rarely occurs early in the course of T1D or before puberty. The recommendation is to perform three tests over a period of 3 to 6 months, two of which need to be positive to confirm the diagnosis of micro albuminuria.
  • In the case of T2D, testing should start at the time of diagnosis of diabetes and repeated annually.

Monitoring: albumin monitoring is recommended to evaluate the effectiveness of therapy, which consists in glycemic control, blood pressure monitoring, dietary protein restriction, and treatment with angiotensin-converting enzyme (ACE) inhibitors. Successful treatment reduces the albumin excretion or diminishes the increase. If diabetes with albuminuria is diagnosed in patients over 75 years, the need for treatment is questionable as the projected life span is too short for nephropathy to develop.

  • T2D: approximately 20–40% of patients with micro albuminuria (30–299 mg/24 h) develop diabetic nephropathy with macro albuminuria ( ≥ 300 mg/24 h), but by 20 years after onset of nephropathy, only 20% will have progressed to end-stage renal failure.

Prognosis: micro albuminuria is a predictive marker for chronic renal failure and increased risk for cardiovascular morbidity and mortality.

  • T1D: in 80% of T1D cases with albuminuria, excretion increases at a rate of 20–30% per year. Patients develop clinical albuminuria (≥ 300 mg/24 h) over a period of 10 to 15 years. In 80% of patients, the glomerular filtration rate then progressively declines over the following years, leading ultimately to end-stage renal failure.


Cardiovascular disease (CVD) is the main cause of morbidity and mortality in type 2 diabetics. To improve this situation, lipid testing must be performed annually. T1D is also associated with an increased risk of CVD. The most common lipid pattern in T2D consists of hypertriglyceridemia and decreased HDL cholesterol /39/. While patients with type 2 diabetes usually have LDL cholesterol levels comparable to those of non diabetics, they have smaller and denser LDL particles which may increase atherogenicity despite normal LDL cholesterol levels. The ADA has defined the following risk categories for the development of CVD /1/:

  • Low risk: LDL cholesterol < 100 mg/dL (2.6 mmol/L), HDL cholesterol > 45 mg/dL (1.15 mmol/L) in men and > 55 mg/dL (1.40 mmol/L) in women, triglycerides < 200 mg/dL (2.3 mmol/L)
  • Medium risk: LDL cholesterol 100–129 mg/dL (2.60–3.50 mmol/L), HDL cholesterol 35–45 mg/dL (0.90–1.15 mmol/L), triglycerides 200–399 mg/dL (2.30–4.50 mmol/L)
  • High risk: LDL cholesterol ≥ 130 mg/dL (3.35 mmol/L), HDL cholesterol < 35 mg/dL (0.90 mmol/L), triglycerides ≥ 400 mg/dL (4.6 mmol/L).

C-peptide, insulin

There are few indications for the routine use of insulin, C-peptide and proinsulin in the diagnosis and monitoring of diabetes. One indication is the determination of insulin in patients with polycystic ovary syndrome, e.g., using the HOMA test. Women with this syndrome have insulin resistance due to hyperandrogenism and disorders of carbohydrate metabolism.

For the initial treatment of T2D, the C-peptide levels are a criterion for selecting the best anti glycemic therapy. However, this does not give rise to the conclusion that, the lower the insulin concentration is prior to therapy, the more appropriate insulin therapy will be /46/. Also refer to the sections on insulin, proinsulin, and C-peptide.


Children with T1D should be tested for TSH and thyroperoxidase (TPO) antibodies following diagnosis to ensure that hypothyroidism is not missed.

Table 3.1-9 Glycemic goals in the treatment of diabetes mellitus /2/

Clinical and laboratory findings

Adult diabetes

Preprandial (pre meal) capillary glucose: 80–130 mg/dL (4.4–7.2 mmol/L).

Postprandial (1–2 h after a meal) capillary glucose: < 180 mg/dL (10.0 mmol/L).

HbA1c: < 7.5%, better 7.0% (DCCT based reference interval 4.0–5.7%).

Lipids: LDL cholesterol < 100 mg/dL (2.6 mmol/L), HDL cholesterol > 50 mg/dL (1.3 mmol/L), triglycerides: < 150 mg/dL (1.71 mmol/L).

Albuminuria: < 30 mg/g creatinine

Blood pressure: ≤ 130/80 mm Hg.

Gestational diabetes

Preprandial (pre meal) capillary glucose: ≤ 95 mg/dL (5.3 mmol/L).

Postprandial (1 h after a meal) capillary glucose: < 140 mg/dL (7.8 mmol/L).

Postprandial (2 h after a meal) capillary glucose: < 120 mg/dL (6.7 mmol/L).

HbA1c < 7% (DCCT based reference interval 4.0–6.0%).

Pregnant diabetic

Preprandial (pre meal), at bedtime and overnight: capillary glucose: 60–99 mg/dL (3.3–5.4 mmol/L).

Peak postprandial capillary glucose: 100–129 mg/dL (5.4–7.1 mmol/L).

HbA1c < 6% (DCCT based reference interval 4.0–6.0%).

Children with T1D

Infants, preschool children

Preprandial (pre meal): capillary glucose: 100–180 mg/dL (5.6–10.0 mmol/L).

At bedtime and overnight: capillary glucose: 110–200 mg/dL (6.1–11.1 mmol/L).

HbA1c < 8.5% but > 7.5% (DCCT based reference interval 4.0–6.0%) because of high risk of hypoglycemia.

Schoolchildren aged 6–12

Preprandial (pre-meal): capillary glucose: 90–180 mg/dL (5.0–10.0 mmol/L).

At bedtime and overnight: capillary glucose: 100–180 mg/dL (5.6–10.0 mmol/L).

HbA1c < 8.0% (DCCT based reference interval 4.0–6.0%), risks of hypoglycemia and diabetic complications before puberty are low.

Adolescents and young adults

Preprandial (pre meal): capillary glucose: 90–130 mg/dL (5.0–7.2 mmol/L).

At bedtime and overnight: capillary glucose: 90–150 mg/dL (5.6–8.3 mmol/L).

HbA1c < 7.5% (DCCT based reference interval 4.0–6.0%), a lower level (< 7%) is desirable, if achievable without hypoglycemia.

With respect to lipids, albuminuria and blood pressure, the same criteria as for adults apply to children from age ≥ 10 years.

Hospitalized patients – Generalized

In non-critically ill insulin-treated diabetics, preprandial glucose levels should be < 140 mg/dL (7.8 mmol/L) and random glucose levels should be < 180 mg/dL (10 mmol/L). Significantly elevated levels can indicate stress-induced hyperglycemia. In these cases, an HbA1c level > 6.5% indicates pre-existing diabetes.

– Critically ill

In critically ill patients, insulin therapy should be started if there is persistent hyperglycemia with levels ≥ 180 mg/dL (10.0 mmol/L). Once insulin therapy has begun, a goal of 140–180 mg/dL (7.8–10.0 mmol/L) is recommended. Glucose concentrations should then be retested every 4 to 6 hours. Levels should not fall below < 110 mg/dL (6.1 mmol/L) /2/. In the NICE SUGAR study /56/ a target blood glucose level of 180 mg/dL (10 mmol/L) was associated with a lower mortality rate than glucose goals in the range of 81–108 mg/dL (4.5–6.0 mmol/L). Severe hypoglycemia with levels < 40 mg/dL (2.2 mmol/L) was recorded in 6.8% of critically ill patients.

– Enteral nutrition

Hyperglycemia is a common side effect of enteral nutrition in hospital. Mean glucose levels should be 160 mg/dL (8.9 mmol/L). Glucose levels should be retested every 4 to 6 hours.

– Parenteral nutrition

The large glucose load provided in parenteral nutrition inevitably leads to hyperglycemia. Insulin therapy is recommended with the same target levels as for critically ill patients. Glucose levels should be retested every 4 to 6 hours.

GlucoCorticoid therapy

Hyperglycemia is a common complication under high-dose glucocorticoid therapy. Glucose levels should be monitored at least within the first 48 h following initiation of therapy, and insulin therapy should be administered if glucose levels are in the diabetic range (Tab. 3.1-11 – Medications that can cause impaired glucose tolerance or hyperglycemia).

Sports and physical exercise

In the case of suboptimal glycemic control with insulin or insulin secretagogues, carbohydrates should be administered to prevent hypoglycemia if glucose levels are < 100 mg/dL (5.6 mmol/L) /2/.

Assessment of treatment goals

In practice, the success of the stated goals of diabetes treatment is suboptimal /57/. Only 57.1% of patients achieve an HbA1c level < 7.0%, only 45.5% a blood pressure < 130/80 mm Hg, only 46.5% cholesterol levels < 200 mg/dL (5.17 mmol/L), and only 12.2% achieve all three stated targets.

Table 3.1-10 Correlation of plasma glucose with the HbA1c level /2/

HbA1c level (%)

Mean plasma glucose
























Table 3.1-11 Drug- or chemical-induced diabetes

Effect on carbohydrate metabolism


Glucokorticoids increase hepatic glucose release, inhibit the insulin-stimulated glucose uptake in the peripheral tissues, increase the concentration of insulin and proinsulin in the circulation, and change the synthesis of insulin in the β-cell, leading to an increased proinsulin-to-insulin ratio in the blood circulation. The receptor and post-receptor functions of the tissues are also changed. Overall, these effects predispose the patient to develop hyperglycemia and make glycemic control difficult in patients with pre-existing diabetes and patients with steroid diabetes. Glucocorticoid-induced hyperglycemia typically manifests as a slight increase in fasting plasma glucose, a marked increase in postprandial glucose, and a decrease in insulin sensitivity. The extent of hyperglycemia depends on the patient’s pre-existing glucose tolerance /60/. Long-term glucocorticoid therapy can lead to steroid diabetes in 50% of cases, depending on the dose. Low doses of < 10 mg prednisone/day rarely cause hyperglycemia. Administration of high doses in the morning leads to marked hyperglycemia in the late afternoon and at night. If diabetes is present, marked hyperglycemia can develop, but ketoacidosis occurs rarely. High doses of glucocorticoids are administered, for example, to neurosurgery patients, cases with cerebral edema or multiple sclerosis, asthmatics, patients with systemic lupus erythematosus, or transplant patients. At the start and during glucocorticoid therapy, blood glucose levels must be measured. After discontinuing glucocorticoids, it can take weeks to months for the induced hyperglycemia to normalize.

Sympathomimetic drugs

Ephedrine, pseudoephedrine, phenyl ephedrine and phenylpropanolamine cause a slight increase in blood glucose in addition to an increase in blood pressure. In children with diabetes, elevated blood glucose concentrations as well as ketonuria following oral intake of such drugs were reported. Administration of albuterol, ephedrine, terbutaline and ritodrine is also reported to cause hyperglycemia and ketonuria. Cough syrup containing ephedrine is reported to have no effect on fasting and postprandial glucose levels in type 2 diabetics. Diabetics who require sympathomimetic drugs should be started on a low dose, and their blood glucose levels should be monitored. Sympathomimetic-induced hyperglycemia is reported to be due to the stimulation of glycogenolysis and gluconeogenesis /81/.


β-blockers can impair glucose tolerance in diabetics and non diabetics. The effect is usually only minor, although cases of severe hyperglycemia have been reported. Selective β1-blockers (atenolol, bisoprolol, metoprolol), which have a cardioselective effect, have a less impairing effect on glucose tolerance than non-selective β1- and β2-blockers (oxprenolol, pindol, propranolol). The effects of β-blockers are dose and time dependent and partly additive to diuretics-induced hyperglycemia. Following the end of therapy, it can take months for the impaired glucose tolerance to return to the pre-therapy condition. The β-blocker-induced impairment in glucose tolerance is thought to be caused by reduced insulin secretion or increased insulin resistance due to the β-blockage /81/.


Diuretics impair glucose tolerance in diabetics and cause or worsen hyperglycemia. However, the effect of the individual diuretics varies:

  • Thiazide-type diuretics (hydrochlorothiazide, thiazide-like diuretics) have the strongest hyperglycemic effect. Hydrochlorothiazide doses of > 25 mg/day increase the likelihood of hyperglycemia.
  • Loop diuretics (furosemide, piretanide, torasemide) have a markedly weaker diuretic effect.
  • Potassium-sparing diuretics (spironolactone, potassium canrenoate) have no effect on glucose tolerance, unless they are used in combination with a thiazide diuretic.

The impairment in glucose tolerance is extremely variable and depends on the dosage and duration of use. The incidence of impaired glucose tolerance is reported to be 10–30% and to occur following the start of treatment: in diabetics within 2 to 4 weeks, in individuals predisposed for diabetes within weeks to months, and in non diabetics after months to years of treatment /81/. The hyperglycemic effect of diuretics is thought to be due to reduced insulin action caused by hypokalemia.

Oral contraceptives (OC) /82/

There is no clear evidence to suggest that estrogen-progestin (EP) OCs increase the risk of diabetes mellitus, even if there is a history of gestational diabetes. Type 1 diabetics should not use OCs if micro- or macro vascular complications are indicated (diabetes for > 20 years, dyslipidemia, hypertension, renal insufficiency, smoking). Since type 2 diabetes is associated with obesity, insulin resistance and cardiovascular risk factors, estrogen-progestin OCs should not be prescribed. Alternatively, non-hormonal or progestin-containing OCs are used.

Nicotinic acid (niacin)

This water-soluble B vitamin is a good lipid-lowering agent, because when taken in therapeutic dosages it reduces the serum concentration of triglycerides by 20–50% and that of low-density lipoproteins by 10–25%. High-density lipoprotein levels are increased by 15–35%. Since type 2 diabetics have hypertriglyceridemia, niacin would be a suitable lipid-lowering agent, but because it causes insulin resistance, hyperglycemia and hyperuricemia in diabetics, it is not a first-choice anti lipemic for these patients /81/.

Table 3.2-1 Hypoglycemia syndromes in adults /1314/

Clinical and laboratory findings

Diabetes mellitus

In non diabetics, a rise in blood glucose levels ≥ 126 mg/dL (7.0 mmol/L) causes a sharp increase in insulin levels, while a decline < 72 mg/dL (4.0 mmol/L) causes insulin secretion to decrease and finally stop. The insulin profile of healthy individuals is characterized by stable levels between meals and sharp, short insulin peaks after meals.

In diabetics the severity of hypoglycemia cannot be defined based on glucose levels, as it varies greatly between patients, being dynamic rather than static. Thus, in poorly controlled diabetics the glucose levels at which hypoglycemic symptoms occur tend to be higher than in non diabetics. Strictly controlled diabetics, in contrast, are more likely to have hypoglycemia and are better at tolerating low glucose levels without developing symptoms. The lower glucose threshold should be individualized by self-monitoring of glucose levels. The American Diabetes Association has defined biochemical hypoglycemia in diabetics as a plasma glucose concentration < 70 mg/dL (3.9 mmol/L) /14/.

In an online diabetes hypoglycemia social network, insulin-dependent diabetics reported frequent minor episodes (low glucose values) in the past 2 weeks and severe episodes (loss of consciousness, seizures, need for glucagon, medical treatment or assistance of another individual) in the past year. 49.1% reported more than 4 episodes of going low in the past 2 weeks, 29.2% had ≥ 1 severe episode, and 16.6% reported both more than 4 recent low episodes and ≥ 1 severe event in the past year /15/.

Type 1 diabetes (T1D) /1314/: hypoglycemia is a significant problem in the daily life of diabetics. On average, diabetics have two episodes of symptomatic hyperglycemia per week. Approx. 2–4% of deaths in these patients are due to severe hypoglycemia.

In T1D, hypoglycemia is the result of the interplay of absolute or relative therapeutic insulin excess in combination with compromised physiological (syndrome of defective glucose counter regulation) and behavioral (syndrome of impaired hypoglycemia awareness) defenses against falling plasma glucose concentrations (Fig. 3.2-3 – Hypoglycemia-associated autonomic failure in insulin-dependent diabetics).

The concept of hypoglycemia-associated autonomic failure (HAAF) posits that recent antecedent iatrogenic hypoglycemia causes /13/:

  • Defective glucose counter regulation by reducing epinephrine responses to a given level of subsequent hypoglycemia in the setting of absent decrements in insulin and absent increments in glucagon
  • Hypoglycemia unawareness by reducing sympathoadrenal and resulting neurogenic symptom responses to a given level of subsequent hypoglycemia.

Many patients have reduced hypoglycemia awareness approximately 5 years following diagnosis of diabetes. HAAF is reversible by 2–3 weeks of scrupulous avoidance of hypoglycemia.

Type 2 (T2D) diabetes /1314/: in the early phase of T2D the counter regulatory mechanisms for preventing hypoglycemia are intact and iatrogenic hypoglycemia only occurs in the insulin-dependent phase, where it is of approximately the same severity as in T1D.

Physical work

Skeletal muscle takes up 90% of the circulating glucose. Endurance athletes often have glucose levels in the range defined as being diagnostic of hypoglycemia. Hypoglycemia causes neuroglycopenic symptoms such as fatigue /15/.

Postprandial reactive hypoglycemia (PRH)

PRH is present if sympathoadrenal and neuroglycopenic symptoms develop concurrently with whole-blood glucose levels < 55 mg/dL (3.0 mmol/L) in clinically healthy individuals. Concomitant insulin levels are at least 3 mU/L and C-peptide levels at least 0.6 μg/L /16/.

Functional hyperinsulinism: characteristics are a delayed, excess insulin response in the absence of insulin resistance. The resulting hypoglycemia leads to increased secretion of catecholamines and cortisol. The diagnosis of an insulinoma must be ruled out by performing a 72-h fast. Neither the oGTT nor mixed-meal tolerance tests are suitable as they frequently give false results /4/. Self-monitoring of blood glucose is considered the gold standard. In a study /12/ of patients with suspected PRH, 46% had whole-blood glucose levels < 60 mg/dL (3.3 mmol/L) and 18% had levels < 50 mg/dL (2.8 mmol/L).

Alimentary hypoglycemia: this form of hypoglycemia is thought to be caused postprandial due to rapid gastric emptying, since the symptoms are common in patients with partial or total gastrectomy /4/. The rapid gastric dumping causes increased secretion of insulin-stimulating gastrointestinal hormones, such as gastrointestinal polypeptide (GIP), and of glucagon-like polypeptide 1 (GLP-1) /17/.

Gastric bypass surgery: in cases of morbidly obese patients, a Roux-en-Y gastric bypass is performed as a weight loss procedure. Following this surgery, some patients experience postprandial dumping syndrome (flushing, weakness, dizziness, blushing, but no neuroglycopenic symptoms) as a result of hyperinsulinemic hypoglycemia. Increased levels of β-cell-trophic polypeptides, such as glucagon-like peptide 1, may contribute to the hypertrophy of the islet cells in these patients /16/.

Renal glucosuria: a high renal loss of glucose can cause an imbalance between glucose production and excretion, leading to hypoglycemia. Because insulin response is dependent on the external glucose load, and a significant portion of the absorbed glucose is excreted, the gluconeogenesis cannot compensate due to insufficient stimulation. About 15% of patients with PRH and renal glucosuria have oGTT-detected hyperinsulinism /4/. Also refer to Section 3.4 – Glucose in urine and extravascular fluids.

High insulin sensitivity /4/: increased insulin sensitivity of the tissues is the most common cause of PRH and is present in 50–70% of cases. These patients are reported to have an impaired balance between insulin sensitivity and insulin secretion due to the loss of homeostatic feedback. Physiologically, high insulin sensitivity is compensated by reduced insulin secretion, but if there is an additional pathogenic event with impaired glucagon secretion, there will be an imbalance /18/.

Impaired counter regulation /4/: some cases of PRH are reported to be caused by deficiencies in glucagon secretion or by glucagon resistance, as seen in some patients with PRH whose postprandial glucagon levels were 2.5 times higher than normal /19/.

Physical constitution /4/

PRH is more common in very lean individuals than in those of normal weight. For example, of 77 Indian volunteers, 22.6% had glucose levels < 60 mg/dL (3.3 mmol/L) /20/. PRH is also commonly seen in people who have lost a lot of weight. An increase in body fat, in particular intraabdominal fat, reduces insulin resistance while a drastic reduction increases it.

Women are generally more prone to develop hypoglycemia after fasting than men. In the 72-h fast, approximately 40% of women had glucose levels < 40 mg/dL (2.2 mmol/L) and a third of these even had levels < 30 mg/dL (1.7 mmol/L) /21/. Moderately overweight women can also develop PRH, especially those with android fat distribution, because compared with normal-weight women they have hyperinsulinism, which is less frequently the case in women with upper body obesity /4/. This is explained by the fact that in android type obesity the fatty acid-induced insulin resistance is lower than in upper body obesity so that an increased amount of glucose is absorbed by adipose tissue in the post absorptive phase. These women complain especially of late-morning hypoglycemia.

Estrogens and progesterone also have an influence on insulin sensitivity. Since it is twice as high in the follicle phase than in the luteal phase, hypoglycemia is more likely to occur in this phase /22/.

Extra pancreatic tumor hypoglycemia, e.g. fibrosarcoma, mesothelioma, hepatoma, hemangiopericytoma

The symptoms of extra pancreatic tumor hypoglycemia (EPTH) or non-islet cell tumor hypoglycemia (NICTH) are due to the presence of mesenchymal, epithelial or hematopoietic tumors with a general size of > 5 cm in diameter. Pronounced hypoglycemia with blood glucose levels up to 20 mg/dL (1.1 mmol/L) can be measured. In doubtful cases, the diagnosis can be confirmed with a 72-h fast. The patho biochemical effects are as follows /23/:

  • Inhibition of hepatic glucose production by inhibition of glycogenolysis and gluconeogenesis
  • Inhibition of lipolysis with reduction of free fatty acids in blood
  • Increased glucose utilization by the muscles, decreased utilization by the tumor.


Alcohol consumption reduces the production of glucose in the liver by inhibition of glycogenolysis and gluconeogenesis /4/. Alcohol-induced hypoglycemia is the result of an increased NADH/NAD ratio when the hepatic glycogen stores are depleted. This suppresses the conversion of lactate to pyruvate, of α-glycerophosphate to dihydroxyacetonphosphate, of glutamate to α-ketoglutarate and some reactions in the citric acid cycle /9/.

Alcohol and diabetes /24/: in diabetics, the consumption of alcohol is one of the main causes of hypoglycemic coma. Even low to moderate amounts of alcohol can lead to hypoglycemia in insulin dependent diabetes or in patients taking oral anti diabetics, especially if there is no food intake. In particular fasting patients and children are at risk of developing alcohol hypoglycemia. Alcohol consumption is reported to reduce gluconeogenesis and cause increased insulin secretion /25/. Diabetics should drink alcohol only with a meal and should not have more than 1–2 glasses of wine or 1–2 beers or 1–2 mixed drinks over a period of 1.5–2 h /24/.

Laboratory findings: hypoglycemia, hypoinsulinemia; lactate, β-hydroxy butyrate and free fatty acids are elevated, the screening test for ketone bodies in urine is positive. Hypoglycemia develops 6–36 h after alcohol consumption in malnourished persons or individuals who have skipped 1–2 meals. Children readily develop alcohol-induced hypoglycemia.


Due to the reduced supply of alanine, gluconeogenesis is reduced. The small glycogen reserve in the liver is not sufficient to keep the blood glucose concentration within the reference interval /9/.

Polycythemia vera

Low plasma glucose levels are measured due to an uneven distribution of glucose between erythrocytes and plasma as a result of excessive erythrocytic glycolysis.


Hypoglycemia as a result of excessive leukocytic glycolysis or increased glycolysis due to expanded hematopoiesis.

Factitious hypoglycemia

Factitious hypoglycemia cannot be differentiated from an insulinoma based on blood glucose levels (see Section 3.7 – Insulin, C-peptide, proinsulin). Factitious hypoglycemia is associated with a low concentration of C-peptide and elevated insulin level.


The chemotherapeutic pentamidine is prescribed for the prophylaxis and treatment of Pneumocystis carinii pneumonia, the treatment of leishmaniosis and in the early stage of Trypanosoma gambiense infection. Approximately 5–14 days following the start of treatment, 6–40% of patients develop hypoglycemia as a result of the cytolytic reaction of islet cells with the release of insulin. The pancreatic destruction process can lead to insulin deficiency in these patients weeks to months later, or in rare cases even just a few days after the hypoglycemia /24/.


Salicylates have a blood glucose-reducing effect. However, high doses are required, which have a toxic effect if treatment continues for a long time. Aspirin taken at a dose of 4–6 g/day reduces blood glucose levels in healthy and diabetic individuals. In T2D, its effect is reported to be based on increased insulin secretion, increased insulin sensitivity and elevated levels of the medicated sulfonylurea. The cause is reported to be the increased release of the sulfonylurea from its protein binding sites and the reduced renal elimination. Non-steroidal antiphlogistics (indometacin, piroxicam) also cause hypoglycemia /24/.


In certain situations, β-blockers cause hypoglycemia in certain patients. The following examples are reported /24/:

  • Hypoglycemia, bradycardia and low blood pressure in neonates, if the mother was administered a β-blocker before the cesarian section
  • Hypoglycemia in T1D, in glaucoma treatment with timolol eye drops.

It is also important to state that non cardioselective β-blockers can aggravate and prolong an existing hypoglycemia.

Table 3.2-2 Hypoglycemia syndromes in childhood and infancy /33/

Clinical and laboratory findings

Neonatal hypoglycemia /26/

Healthy newborns: in newborns the glucose concentration reaches a physiological nadir by 1–3 hours of age. The 10th percentile is 36 mg/dL (2.0 mmol/L). After the 3rd hour of age blood glucose stabilizes over the following 2 days and 95% of values are > 40 mg/dL (2.2 mmol/L) and 99% above 36 mg/dl (2.0 mmol/L). Healthy overweight newborns with concentrations at the lower end of the reference interval do not have any abnormality when retested at age 4 years.

Neonatal hypoglycemia: acute symptoms include irritability, tremor, lethargy, apnea, sucking weakness, muscular hypotension, hypothermia, high pitched crying, cerebral seizures. Damage to the basal ganglia and the white matter of the occipital cortex is possible if levels are < 27 mg/dL (1.5 mmol/L).

Newborns of diabetic mothers: as a preventive measure, early feeding in the delivery room 30 min. after birth is recommended, and glucose levels should be determined 2 hours later. After 2–3 h, levels should be > 45 mg/dL (2.5 mmol/L). In normal newborns, glucose levels may be 36–45 mg/dL (2.0–2.5 mmol/L) whereas in newborns with antecedent hypoglycemia or clinical symptoms they should always be > 45 mg/dL (2.5 mmol/L). Further postprandial tests 6, 12, and possibly 24 and 48 hours after feeding are recommended. If concentrations are persistently < 30 mg/dL (1.7 mmol/L), intravenous administration of glucose is recommended.

Miscellaneous: hypoglycemia lasting longer than 48–72 h may suggest a disease such as congenital metabolic disorder or hormonal disorder. Despite a high glucose infusion rate, these children will not develop adequate lipolytic activity (free fatty acids < 0.8 mmol/L) or a ketogenic response (β-hydroxy butyrate < 0.1 mmol/L) /25/.

Idiopathic transient hyperinsulinism, Beckwith-Wiedeman syndrome and large tumors are rare causes, the latter causing hypoglycemia through increased secretion of insulin-like growth factors.

Congenital hyperinsulinism caused by gene mutations: ABCC8 (autosomal recessive and dominant), KCNJ11 (autosomal recessive and dominant), GLUD1 (dominant), GCK (dominant), HADH (recessive), HNF4A (dominant), SLC16A1 (motion-induced) (dominant).


Glycogenoses (glycogen storage diseases, GSD), with the exception of glycogen synthase deficiency (GSD 0), are characterized by a lack of glycogen breakdown due to enzyme defects /27/. Hypoglycemia is associated with: GSD 1 (glucose-6-phosphatase deficiency) (GSD 1a), glucose-6-phosphate-translocase deficiency (GSD 1b), amylo-1,6-glucosidase deficiency (GSD 3), liver phosphorylase deficiency (GSD 6 and GSD 9) and GSD 0. While GSDs 1, 3, 6 and 9 have similar clinical presentations, GSD 1 is the most severe type of these four conditions. The main symptoms are failure to thrive, hepatomegaly without splenomegaly and hypoglycemia. Patients, especially those with GSD 1, usually have reduced fasting tolerance of only 2–4 h. Additionally they may have polymorphonuclear granulocyte dysfunction and inflammatory bowel disease. Also refer to the Sections on lactate (Section 5.6) and uric acid (Section 5.4).

Laboratory findings: fasting hypoglycemia and lactate elevated in GSD 0, 1, 6, 9. Ketone bodies reduced in GSD 1 and elevated in GSD 0, 6, 9. Cholesterol, triglycerides and uric acid are elevated in GSD 1.

Fatty acid oxidation disorder (also refer to the Section on uric acid, Section 5.4)

These disorders primarily involve defects of the enzymes long-chain-acyl-CoA dehydrogenase (LCAD), medium-chain-acyl-CoA dehydrogenase (MCAD) and short-chain-3-hydroxy acyl-CoA dehydrogenase (SCAD) /28/. These three enzymes facilitate the mitochondrial breakdown of long-chain fatty acids into ketone bodies via acyl-CoA-intermediates through β-oxidation. The most commonly occurring MCAD defect has an incidence of 1 : 23,000 in Northern Europe; the SCAD defect is very rare. Clinical manifestations of these disorders include symptoms such as reduced food intake, seizures and impaired consciousness following ordinary infections /29/.

Laboratory diagnostics: hypoglycemia, reduced ketone bodies, acidosis, elevated ammonia, elevated lactate in the presence of LCAD defect, elevated uric acid in the presence of MCAD defect, elevated liver enzymes in the presence of LCAD and MCAD defects, elevated CK with LCAD defect in childhood. Few data are available for the SCAD defect /31/.

Disorders of fructose metabolism

All disorders of fructose metabolism are associated with hypoglycemia /27/:

  • Fructokinase deficiency, an asymptomatic disorder
  • Fructose-1-phosphate aldolase deficiency, also known as hereditary fructose intolerance (HFI). Symptoms occur following ingestion of fructose. Diagnosis relies on a fructose tolerance test. Laboratory findings include elevated amino transferases, elevated uric acid, and proteinuria.
  • Fructose-1.6-diphosphatase deficiency. This is a disorder of gluconeogenesis. Children with this disorder are lethargic and have moderate hepatomegaly. Laboratory findings: lactate, ketone bodies and alanine are elevated. Uric acid levels are slightly elevated or normal.
  • D-glycerate-acidemia. Mainly manifests as neurological disorders, although many patients are asymptomatic.


The primary source of galactose is the lactose (milk sugar) ingested through food. There are three known disorders of galactose metabolism: galactokinase deficiency (GALK), galactose-1-phosphate-uridyltransferase deficiency (GALT), and uridine-diphosphate-galactose-4-epimerase deficiency (GALE). Among these three, classic galactosemia (GALT) is the most common and severe type /27/. GALT is an autosomal recessive disorder with an incidence of 1 in 40,000 to 100,000 newborns. Most cases are detected through Guthrie or Weidemann screening tests of newborns. Diagnosis is based on determining galactose and galactose-1-phosphate levels in blood. In two-thirds of cases the disease is acute. Symptoms first appear at the end of the first week of life and occur in the form of vomiting, jaundice, lethargy, food refusal, hypoglycemia and weight loss, followed by the development of hepatomegaly and cataract. A galactose-free diet improves symptoms within 24 h. If galactosemia is present, there is no conversion of galactose-1-phosphate to UDP galactose and glucose-1-phosphate due to the GALT deficiency (Fig. 3.2-4 – Galactose metabolism). As a result of the metabolic block, galactose and galactose-1-phosphate accumulate and are metabolized via alternative pathways. In addition, hypoglycemia develops. The accumulated galactose is either reduced to galactitol or oxidized to galactonic acid. Both products can be detected in urine. Galactitol is reported to be responsible for the development of cataracts, and the accumulated galactose-1-phosphate for other clinical manifestations such as hepatomegaly.

Carnitine shuttle disorder (refer to Section 5.3 – L-Carnitine/28/

The carnitine shuttle transports long-chain fatty acids into the mitochondrion. Rare disorders associated with hypoglycemia are carnitine uptake defects, carnitine palmitoyltransferase deficiency and carnitine-acylcarnitine carrier defect. While individuals with these disorders have elevated ketone bodies and acidosis, this is not the case with carnitine palmitoyltransferase-2 deficiency.

Hormone deficiency

A lack of insulin counter regulatory hormones such as glucagon, growth hormone and cortisol can cause symptoms similar to those of gluconeogenesis defects.

Various causes

Hypoglycemia can occur, e.g., with viral hepatitis, sepsis, hemorrhagic shock, encephalopathy, panhypopituitarism and diarrhea in combination with liver dysfunction.

Hypoglycemia in diabetes type 1

If the infant has minimal clinical symptoms and a normal level of consciousness, it is recommended that capillary blood glucose levels in the range 70–55 mg/dL (3.9–3.0 mmol/L) be classified as mild hypoglycemia, and levels < 55 mg/dL (3.0 mmol/L) as moderate hypoglycemia /30/. Every glucose concentration < 70 mg/dL (3.9 mmol/L) which is accompanied by a reduced level of consciousness or seizures is defined as severe hypoglycemia. The latter mainly occurs at night. 10% of children with type 1 diabetes have nocturnal glucose levels between 70 and 55 mg/dL (3.9–3.0 mmol/L), regardless of the number of daily insulin injections /31/. The prevalence of hypoglycemic events with clinical symptoms in adolescent children is 9.7 episodes annually per 100 patient years /30/.

Table 3.2-3 Laboratory findings and suggestion of disorder in childhood hypoglycemia


Suggestion of disorder

Hypoglycemia, hypoketonemia, low fatty acids

Hyperinsulinism, hypopituitarism

Hypoglycemia, hypoketonemia, elevated fatty acids

Fatty acid oxidation defect (various acyl-CoA-dehydrogenase deficiencies), carnitine metabolism disorder, ketogenesis defect (β-hydroxy-β-methylglutaraciduria)

Hypoglycemia, hyperketonemia, no elevated lactate level

Ketotic hypoglycemia, hypopituitarism (infancy), type 3 glycogenosis (glycogen synthase deficiency)

Hypoglycemia, elevated lactate level

Gluconeogenesis defect such as type I glycogenosis, fructose-1,6-diphosphatase deficiency, pyruvate carboxylase deficiency

Hypoglycemia, hypoketonemia, elevated lactate level

Type I glycogenosis

Hypoglycemia, hyperketonemia, elevated lactate level

Fructose-1,6-diphosphatase deficiency

Table 3.3-1 Blood glucose specimen for evaluation of the metabolic state /7/

Fasting glucose

Blood collection in the morning (e.g., 8 a.m.) after 8–10 h fast.

Preprandial glucose

Blood collection prior to meal, e.g. 30–60 prior to insulin dose.

Postprandial glucose

Blood collection during the postprandial period. The latter usually lasts 4 h, sometimes up to 6 h, and is followed by the postabsorptive period. For practical reasons, blood is drawn 2 h after the start of the meal. In diabetics with a tendency to develop (nocturnal) hypoglycemia, levels can be measured again at 5 a.m. during the postabsorptive state.

Blood glucose day profile

Combination of fasting glucose, preprandial and postprandial glucose, samples taken at, e.g., 8 a.m., 11 a.m., 2 p.m.

Random glucose

Blood samples taken during the day or night, regardless of meals.

Table 3.3-2 Blood glucose reference intervals

Neonates /11/

Cord blood

63–158 (3.5–8.8)

1 h

36–99 (2.0–5.5)

2 h

39–89 (2.2–4.9)

5–14 h

34–77 (1.9–4.3)

20–28 h

46–81 (2.6–4.5)

44–52 h

48–79 (2.7–4.4)

Children, fasting* /12/

60–99 (3.5–5.5)

Adults, fasting** /12/

60–95 (3.3–5.3)

* Data for capillary blood (heel or finger pad), venous and capillary plasma. The values for neonates and children are the 2.5th and 97.5th percentiles.

** Intervals are according to the recommendations of the American Diabetes Association.

For each decade of life the glucose concentration in plasma or whole blood increases by about 2 mg/dL (0.1 mmol/L).

Data are expressed in mg/dL (mmol/L). Conversion: mg/dL × 0.05551= mmol/L

Table 3.3-3 Blood glucose in diabetes mellitus and other categories of glucose regulation

Clinical and laboratory findings

Diabetes mellitus /17/

The American Diabetes Association (ADA) has recommended the following criteria for the diagnosis of diabetes: corresponding clinical symptoms, random hyperglycemia ≥ 200 mg/dL (11.1 mmol/L) in venous plasma, or fasting glucose ≥ 126 mg/dL (7.0 mmol/L), a 2-h level ≥ 200 mg/dL (11.1 mmol/L) following oral glucose load (see Section 3.5 – Oral glucose tolerance test) or HbA1c ≥ 6.5% (48 mmol/mol). If one of these three results is present, a confirmatory test must be performed on a subsequent day. Also refer to Section 3.1.1 – Diabetes mellitus.

Type 1 diabetes: T1D is mainly diagnosed in childhood. Blood glucose levels at diagnosis are commonly 200–400 mg/dL (11.1–22.2 mmol/L). There is increased osmotic diuresis as a result of glucosuria of several grams/day. The test for ketone bodies in urine is positive, usually there is no acidosis. If ketoacidosis is present, blood glucose is usually ≥ 300 mg/dL (16.7 mmol/L), pH is < 7.25, and standard bicarbonate is < 15 mmol/L. Metabolic control is monitored by determining pre- and postprandial blood glucose and the HbA1c value. To prevent nocturnal hypoglycemia, glucose levels may have to be measured at bedtime and again at about 5 a.m.. Capillary blood glucose at bedtime should not be below 120 mg/dL (6.7 mmol/L), and 5 a.m. levels should be above 60 mg/dL (3.3 mmol/L).

Type 2 (T2D) diabetes: the ADA recommends fasting plasma glucose screening of all adults over 45 years with a body mass index > 25 kg/m2. If levels are < 100 mg/dL (5.6 mmol/L), testing should be repeated every 3 years. Glucose levels at diagnosis of T2D are generally lower than in T1D and patients usually have glucosuria, but not ketonuria. Dietary metabolic control or oral antidiabetic treatment is monitored by measuring fasting glucose and/or postprandial glucose concentrations 1–2 hours after a meal. Glycemic goals are blood glucose levels of 70–130 mg/dL (3.9–7.2 mmol/L) preprandial and < 180 mg/dL (10.0 mmol/L) postprandial without glucosuria.

Gestational diabetes

In pregnant women, fasting and preprandial glucose levels should be ≤ 95 mg/dL (5.3 mmol/L) in capillary whole blood and ≤ 92 mg/dL (5.1 mmol/L) in plasma (see Section 3.3-5 – Clinical significance).

Diabetic coma

There are two types of diabetic coma: keto acidotic coma and hyper osmolar non- ketotic syndrome. The latter most frequently affects older individuals with type 2 diabetes (see Section 5.5 – Ketone bodies).

Keto acidotic coma: laboratory findings are plasma glucose levels > 300 mg/dL (16.7 mmol/L), standard bicarbonate < 15 mmol/L, pH < 7.25, serum osmolality > 340 mmol/kg, ketonuria, and serum β-hydroxy butyrate levels > 5 mmol/L. The coma may present with leukocytosis as well as increased lipase and α-amylase activities and acute abdomen.

Hyper osmolar non-ketotic syndrome: laboratory findings are severe hyperglycemia with glucose levels > 600 mg/dL (33.3 mmol/L), elevated serum osmolality, but lack of ketoacidosis.

Acute myocardial infarction (AMI) /18/

Elevated blood glucose levels at the time of an AMI are reported to be associated with arrhythmia, larger size of infarction and reduced re perfusion. In addition, it is reported that the coagulation system is activated and consequently the risk of thrombotic events is increased. This should be the result of the reduced half life of fibrinogen and elevated concentrations of fibrinopeptide A, prothrombin fragments and factor VII. Hyperglycemia is also reported to have a pathogenic effect due to the generation of free oxygen radicals. Overall, the risk of mortality is higher in AMI patients who have hyperglycemia at the time of hospital admission.

Acute stroke

Focal cerebral ischemia combined with hyperglycemia leads to more severe tissue damage and larger infarction of the brain than with normoglycemia. A study /19/ reported a larger infarct volume with glucose levels > 180 mg/dL (9.9 mmol/L) than with lower levels. Hyperglycemia at admission is also a predictor for higher long-term morbidity and mortality /20/.

Craniotomy /21/

Dexamethasone is often administered during craniotomy in order to minimize cerebral edema and damage. Patients who were given dexamethasone during and after the surgery but not before, had the highest peak blood glucose levels (198 ± 36 mg/dL; 11.0 ± 2.0 mmol/L; mean ± SD) compared to those who were not given dexamethasone (141 ± 38 mg/dL; 7.8 ± 2.1 mmol/L) and those who were administered dexamethasone before and during the surgery (153 ± 22 mg/dL; 8.5 ± 1.2 mmol/L).

Dementia /22/

The Adult Changes in Thought (ACT) study examined the relationship between blood glucose and the risk of dementia in 2067 individuals with an average age of 76 over a 6.8-year observation period. Approximately 11% of the participants had diabetes. Among the participants without diabetes, those who did not develop dementia had average glucose levels of 100 mg/dL (5.5 mmol/L) while those who did develop dementia had average levels of 115 mg/dL (6.4 mmol/L). Among the participants with diabetes, the corresponding glucose levels were 160 mg/dL (8.9 mmol/L) and 190 mg/dL (10.5 mmol/L). 25.4% of the participants developed dementia. The mean hazard ratio for dementia was 1.18 for non diabetics and 1.40 for diabetics.

Hepatitis C (HCV) /23/

In patients with HCV infection, hyperglycemia is often associated with insulin resistance or diabetes. These patients can be divided into two groups: those with HCV-induced hyperglycemia and those who have prediabetes or diabetes and unrelated HCV infection. Patients whose hyperglycemia was induced by HCV usually have a family history of diabetes. They differ from patients with classic T2D in that they have a lower body mass index and lower cholesterol and LDL levels.

Parenteral nutrition /24/

Hyperglycemia develops in 10–88% of patients on total parenteral nutrition. Blood glucose levels above 180 mg/dL (10.0 mmol/L) 24 hours before the start of treatment are associated with increased mortality and complications such as pneumonia during hospitalization.

Neonates /25/

Neonatal hyperglycemia is usually defined as blood glucose level ≥ 126 mg/dL (7.0 mmol/L). Hyperglycemia is common during the first weeks of life of infants born more than 12 weeks premature. The pathophysiology of hyperglycemia is due to insulin resistance of the liver associated with increased glucose production, as well as insulin resistance of peripheral tissues. Such glucose levels are rare in full-term infants. Often pre term infants are treated with glucose infusions, which exacerbate hyperglycemia. Adverse effects of neonatal hyperglycemia include osmotic diuresis, dehydration and weight loss. The consequences are increased mortality and increased incidence of intracranial hemorrhage. These adverse effects are reported to only occur with blood glucose levels > 270 mg/dL (15 mmol/L) or urine glucose excretion > 20 g/L.

Table 3.3-4 Clinical significance of fasting glucose and postprandial blood glucose

Fasting glucose

Impaired fasting glucose (IFG) levels of 100–125 mg/dL (5.6–6.9 mmol/L) are indicative of reduced insulin secretion. This is the case with prediabetes and in the early phase of type 2 diabetes (T2D).

Postprandial glucose

Elevated postprandial glucose levels are a sign of insulin resistance. This is the case if the 2-h plasma glucose level is 140–199 mg/dL (7.8–11.1 mmol/L). during OGTT. Insulin resistance is an independent predictor of T2D. Also refer to Section 2.2 – Metabolic syndrome.

Table 3.3-5 Methodological errors in glucose measurement


Clinical and laboratory findings

Hexokinase method

Specific method for glucose determination. If the assay is to be performed on a deproteinized sample, trichloroacetic acid cannot be used as it inhibits glucose-6-phosphate dehydrogenase. Perchloric acid is also unsuitable, because arsenates, phosphates and uric acid are not removed; in addition, neutralization or an increased buffering capacity would be required. If glucose levels are measured in the hemolysate, maleinimide has to be added to inhibit erythrocyte enzymes. No interfering drugs are known.

Glucose dehydrogenase (Gluc-DH method)

Of the sugars normally occurring in blood, glucose and xylose are measured. Xylose levels are usually < 2.5 mg/dL and therefore not a problem. During and after a xylose load test the glucose dehydrogenase method cannot be used to determine blood glucose levels. No interfering drugs are known.

Glucose oxidase method

False low results are produced by erythrocyte glutathione when capillary or whole blood hemolysate is analyzed.

Test strips for blood glucose quick tests and for patient self-monitoring

The commercial test devices available measure glucose concentrations between 20 to 800 mg/dL (1.1 to 44.4 mmol/L). Some of the glucose meters provide acceptable results when used by an experienced person. The systems should not be used when accurate glucose measurements are required, e.g. in the oGTT or for the diagnosis of hypoglycemia. The results obtained with glucose meters are too low with Hct values > 55% and too high with Hct values < 35%. Glucose meters have been shown to provide satisfactory results for patient self-monitoring /37/. These systems can be recommended for self-monitoring of blood glucose (SMBG) /38/.


Most analyzers measure glucose in diluted samples. Electrochemical biosensors, however, can measure glucose activity directly in the aqueous phase of the sample. The activity equals the molality, the activity coefficient being close to 1.0. Analyzers with direct-reading biosensors are calibrated with aqueous calibrators in order to measure the concentration of glucose in relation to its molality in the patient sample. The ratio of glucose measured by this method to glucose measured in a diluted sample is 1.18 for whole blood and 1.06 for plasma. Glucose concentrations determined with biosensors are higher compared to those measured with methods involving sample dilution. It is recommended that the results from direct-reading biosensor glucose analyzers be converted to plasma glucose (Fig. 3.3-1 – Conversion factors for glucose/29/. The analyzers are programmed do this automatically.

Table 3.4-1 Glucose reference intervals in body fluids

  • Spontaneous urine: up to 165 mg/L (0.92 mmol/L) /2/
  • CSF: 48–76 mg/dL (2.7–4.2 mmol/L)
  • Puncture fluid: 74–106 mg/dL (4.1–5.9 mmol/L)

Conversion: mg/l × 0.00555 = mmol/L.

Table 3.4-2 Diseases and conditions associated with glucosuria

Clinical and laboratory findings

Diabetes mellitus

Type 1 diabetics always have glucosuria in the range of several g/24 h at the time of diagnosis, whereas this is often not the case with type 2 diabetics.

Urinary glucose determination is no longer relevant for the metabolic control of diabetics. Instead, the focus is on self-monitoring of blood glucose /4/.

Renal glucosuria

Renal diabetes is a disorder of glucose reabsorption. It can occur as an inherited disorder and associated with the tubular reabsorption disorder of other substances, and as an acquired disorder with kidney diseases. The daily glucose excretion can be > 50 g with normal blood glucose levels. Renal glucosuria can be differentiated from the diabetic type by the following findings: glucosuria with normoglycemia, normal glucose tolerance test, normal HbA1c value, reduced fractional extraction of glucose FEG (%) /5/.

Toxic kidney damage

Glucosuria can occur in the presence of toxic kidney damage and chronic kidney diseases with damage to the proximal tubules.

Gestational glucosuria

During pregnancy, the renal threshold for glucose is lowered. If blood glucose levels are normal, glucose excretion in urine usually increases after the 3rd month of pregnancy. Excretion is highest in the last trimester and is not accompanied by excretion of ketone bodies. Glucosuria in pregnancy should always be investigated in order to rule out manifest diabetes or gestational diabetes. A reduced FEG (%) for glucose is an indicator for gestational glucosuria /6/.

Table 3.4-3 Diagnostic significance of glucose measurements in CSF and extravascular fluids


Clinical and laboratory findings

Cerebrospinal fluid (CSF)

Bacterial infection and/or cell proliferation, in particular diseases with neutrophil granulocytosis, are associated with reduced glucose levels. In the presence of bacterial meningitis the CSF/blood glucose ratio is < 0.5. The decrease in glucose and the concomitant increase in CSF lactate are not due to the bacteria or granulocytes but are caused by the anaerobic metabolism of the brain and the impaired glucose transport across the blood/brain barrier /7/.


In the presence of non-bacterial peritonitis the ascites/blood glucose ratio is ≥ 1.0, with bacterial peritonitis it is < 1.0. Patients with peritoneal carcinosis are also reported to have a lower ratio /8/.

Pleural fluid

The glucose concentration corresponds to that in blood /9/. Reduced levels with pleural fluid/blood glucose ratios < 1.0 can occur in the presence of bacterial, tuberculous, malignant and rheumatoid effusions.

Table 3.4-4 Interferences in urine glucose determination using the glucose oxidase method


Clinical and laboratory findings

Urine pH < 5

The enzymatic detection reaction is slowed down. Urine pH levels < 5 occur with high concentrations of acetoacetic acid, β-hydroxybutyric acid and nalidixic acid. If the reaction is negative but there is clinical suspicion of glucosuria, a test for ketone bodies in urine must be performed.

Ingestion of

  • Ascorbic acid
  • Salicylic acid

(> 2 g per day)

Compounds that have a lower redox potential than the indicator dye interfere with the indicator reaction of the test strips. The main interfering factors are ascorbic acid and gentisic acid, the latter being a metabolite of salicylic acid. In a study /10/, 3–20% of urine samples had ascorbic acid concentrations of 100–200 mg/L (0.6–1.2 mmol/L). Higher concentrations were rare. Test strips that use potassium iodide chromogen as an indicator system show falsely low results only in the presence of high ascorbic acid concentrations. In contrast, if test strips containing tetra methyl benzidine are used, glucosuria of 500 mg/L (2.78 mmol/L) is almost completely suppressed by ascorbic acid in a concentration of 200 mg/L (1.2 mmol/L).


Residues of peroxides from cleaning agents in the containers lead to false positive results or false high glucose results.

Table 3.5-1 Diagnosis of diabetes mellitus and impaired glucose tolerance using 75 g oGTT /67/


Whole blood


Blood collection





Diabetes mellitus

Fasting level


≥ 126 (7.0)

≥ 126 (7.0)


2-h level

≥ 180 (10.0)

≥ 200 (11.1)

≥ 200 (11.1)

≥ 220 (12.2)

Impaired glucose tolerance (IGT)

2-h level

≥ 120 and < 180 (≥ 6.7 and < 10.0)

≥ 140 and < 200 (≥ 7.8 and < 11.1)

≥ 140 and < 200 (≥ 7.8 and < 11.1)

≥ 160 and < 220 (≥ 8.9 and < 12.2)

Fasting glucose

< 90 (5.0)

< 100 (5.6)

< 100 (5.6)

< 110 (6.1)

Impaired fasting glucose (IFG)


100–125 (5.6–6.9)

100–125 (5.6–6.9)


Data in mg/dL (mmol/L). The levels for venous whole-blood hemolysate and capillary plasma are not based on the WHO recommendation.

Table 3.5-3 Criteria of 75 g oGTT depending on specimen selection /15/


Whole blood







≥ 85 (4.7)

≥ 90 (5.0)

≥ 92 (5.1)

≥ 95 (5.3)


≥ 165 (9.2)

≥ 180 (10.0)

≥ 180 (10.0)

≥ 200 (11.1)


≥ 140 (7.8)

≥ 155 (8.6)

≥ 153 (8.5)

≥ 170 (9.4)

Data in mg/dL (mmol/L). The venous plasma levels are based on the recommendations of the International Association of Diabetes and Pregnancy Study Groups (see Tab. 3.1-6 – Testing for diabetes in pregnant women). The levels for the other sample materials have been adapted.

Table 3.5-2 Factors influencing glucose tolerance

Hyperlipoproteinemia, liver cirrhosis, metabolic acidosis (uremia), long-term confinement to bed, hyperthyroidism, pregnancy, potassium deficiency, severe heart failure, starvation, stress (heart attack, surgery, other trauma).

Treatment with saluretics (in particular thiazides), corticosteroids, laxatives, nicotinic acid, nitrazepam, phenothiazines, phenacetin, thyroid hormones, non-steroid antiphlogistics, use of oral contraceptives.

Table 3.5-4 Influencing factors that can cause a false pathological result in the oGTT


Clinical and laboratory findings

Duodenal ulcer

In these patients, glucose loading leads to increased secretion of small intestine hormones which induce hyperglycemia.

Condition after Billroth II surgery

Accelerated glucose reabsorption, since glucose is transported into the intestine faster and in larger amounts, leading to higher and earlier blood glucose spikes than normal. In these patients the oGTT cannot be performed.

Hypokalemia, hypomagnesemia

Glucose tolerance tests should only be performed if potassium and magnesium levels are normal. Hypokalemia and hypomagnesemia can simulate a diabetic metabolic condition.

Insufficient carbohydrate intake

This situation can arise in patients who have been in a fasting state for a prolonged period of time prior to oGTT.

Oral contraceptives

Oral contraceptives, in particular combinations of estrogens and gestagens, may give rise to pathological carbohydrate tolerance.


Diuretics (sulfonamide derivatives, ethacrynic acid) and laxatives cause hypokalemia and can thus impair glucose tolerance; they should be discontinued 1 week prior to testing.

Table 3.5-5 Influencing factors that can lead to a false normal oGTT result

Clinical and laboratory findings


In patients with acute enteritis, regional enteritis, irritable colon, ulcerative colitis, glucose-galactose intolerance, disaccharidase deficiency, Whipple’s disease, tuberculosis, or parasite infestation, a large portion of the orally ingested glucose is not resorbed. As a result, plasma glucose levels increase only slightly or not at all. Patients often develop diarrhea in response to the glucose drink.


Caffeine, reserpine, biguanides, monoamine oxidase inhibitors, blood glucose reducing sulfonamide derivatives, gonadotropin and medium-chain fatty acids improve carbohydrate tolerance, partly by stimulating the release of insulin.

Physical activity

Physical activity during the 2-h oGTT or a high room temperature which causes the patient to sweat can increase glucose metabolism through the secretion of catecholamines and thus lead to a false normal oGTT result.

Fluid intake

Increases the motility and secretion of the gastrointestinal tract and reduces glucose absorption.

Table 3.6-1 Standard interpretation of HbA1c




Reference interval



Low risk of diabetes

< 40

< 5.8

Increased risk of diabetes



Diabetes mellitus

> 46

> 6.4

Therapeutic target



Change of therapy required



* Acc. to Diabetes Control and Complications Trial (DCCT) 1993, American Diabetes Association (ADA) 2010, 2011

Table 3.6-2 HbA1c and hyperglycemia /18/

Clinical and laboratory findings

Screening for diabetes

The American Diabetes Association (ADA) recommends screening all adults over 45 years of age as well as all individuals, regardless of age, with a body mass index over 25 kg/m2 who have one or more of the risk factors listed in Tab. 3.1-2 – Diagnosis of prediabetes and diabetes mellitus based on ADA criteria for the presence of diabetes. This can be done using fasting plasma glucose (FPG), the oGTT, and HbA1c /8/. If results are normal, screening should be repeated every 3 years. HbA1c was added as a test criterion, since FPG is unreliable due to diurnal and postprandial variations and its instability in plasma as a result of glycolysis.

Categories with an increased risk for diabetes mellitus (prediabetes)

Prediabetes is the term used for individuals whose glucose levels do not meet the criteria for diabetes but are too high to be considered normal /2/. The diagnostic criteria are /22/:

  • Fasting glucose 100–125 mg/dL (5.6–6.9 mmol/L)
  • 2-h oGTT plasma glucose 140–199 mg/dL (7.8–11.0 mmol/L)
  • HbA1c 5.7–6.4% (39–47 mmoL/moL).

All three tests are associated with an increased risk which continues at the lower end and increases disproportionally at the upper end. Individuals at increased risk should lose 5–10% of their body weight to prevent type 2 diabetes. They are also at increased risk of developing cardiovascular disease.

Several prospective studies that used HbA1c to predict the progression to diabetes as defined by HbA1c criteria demonstrated a strong, continuous association between HbA1c and subsequent diabetes. In a follow-up interval averaging 5.6 years (range 2.8–12 years), those with HbA1c between 5.5 and 6.0% (between 37 and 42 mmol/mol) had an increased risk of diabetes (5 year incidence from 9 to 25%). Those with a HbA1c range of 6.0–6.5% (42–48 mmol/mol) had a 5-year risk of developing diabetes between 25 and 50% and a relative risk 20 times higher compared with HbA1c of 5.0% (31 mmol/mol) /2/.

Diagnosis of diabetes mellitus

The diagnosis of diabetes mellitus is based on confirming the presence of hyperglycemia. This can optionally be /22/:

  • Random glucose levels ≥ 200 mg/dL (11.1 mmol/L)
  • Fasting plasma glucose after at least 8 hours of fasting ≥ 126 mg/dL (7.0 mmol/L)
  • Plasma glucose 2 hours after 75-g oGTT ≥ 200 mg/dL (11.1 mmol/L)
  • As a new criterion, an HbA1c result of ≥ 6.5% (48 mmol/mol). Confirmation by repeat testing on a subsequent day, unless there are additional clinical symptoms of hyperglycemia, or plasma glucose levels are ≥ 200 mg/dL (11.1 mmoL/L).

Confirming diagnosis /2/

  • Unless there is a clear diagnosis (e.g., patient in a hyperglycemic crisis or with classic symptoms of hyperglycemia and a random plasma glucose ≥ 200 mg/dl; 11.1 mmol/l), a second test is required for confirmation. It is recommended that the same test be repeated or a different test be performed without delay using a new blood sample for confirmation.
  • For example, if the HbA1c result is 7% (53 mmol/mol) and a repeat result is 6.8% (51 mmol/mol), the diagnosis of diabetes is confirmed.
  • If two different tests (such as fasting glucose and HbA1c ) are both above the diagnostic threshold, this also confirms diagnosis.
  • If the patient has discordant results from two different tests, than the result that is above the diagnostic threshold should be repeated, with consideration of the possibility of HbA1c assay interference. The diagnosis is made based on the confirmed test. For example, if a patient meets the diabetes criterion of HbA1c (two results ≥ 6.5% ; 48 mmol/mol) but not fasting glucose ≥ 126 mg/dL (7.0 mmol/l), that person should nevertheless be considered to have diabetes.
  • Because of potential pre analytic variability a normal result is likely for fasting plasma glucose and 2-h plasma glucose in the oGTT. In such cases the health care professional should follow the patient closely and repeat the test.

Glycemic control in adult diabetics

The goal is an HbA1c number below 7.5%, preferably 7.0%. With this goal there is a significant reduction in microvascular complications /22/.

Pregnant women

Determination of HbA1c will not replace 75-g oGGT for hyperglycemia in pregnant women.

Pregnant women with pre-existing diabetes

The goal for glycemic control should be an HbA1c number below 6%. Preconceptional control should be as tight as possible, with a goal of less than 6% /22/.

Glycemic control in children and adolescents

Because of new evidence regarding the risks and benefits of tight glycemic control the ADA recommends a HbA1c goal below 7.5% (Diabetes Care 2015; 38 (Suppl. 1) /22/.

Risk of diabetes, cardiovascular disease and stroke

The Atherosclerosis Risk in Communities (ARIC) study determined the risk of diabetes, cardiovascular disease and stroke as a function of the HbA1c number as a hazard ratio for non diabetics. The hazard ratios (stated in brackets) for the development of diabetes and for cardiovascular disease and stroke were dependent on the HbA1c ranges /23/:

  • Diabetes: < 5.0% (0.52); 5.0 to ≤ 5.5% (1.0); 5.5 to ≤ 6.0% (1.86); 6.0 to ≤ 6.5% (4.48) and ≥ 6,5% (16.47)
  • Cardiovascular disease and stroke: < 5.0% (0.96); 5.0 to ≤ 5.5% (1.0); 5.5 to ≤ 6.0% (1.23); 6.0 to ≤ 6.5% (1.78) and ≥ 6,5% (1,95).

Chronic kidney disease (CKD) /24/

The association of HbA1c with the risk of CKD was investigated by the Atherosclerosis Risk in Communities (ARIC) study over a period of 14 years. Compared to an HbA1c below 5.7%, the hazard ratios (HRs) for CKD were 1.12 for the HbA1c range of 5.7–6.4% and 1.39 for the range ≥ 6,5%. The corresponding HRs for end stage renal disease were 1.51 and 1.98.

Mortality risk in type 2 (T2D) diabetes

The epidemiologic relationship between the HbA1c goal and total mortality was examined by the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial in 10,251 diabetics with an average age of 62 over a period of 3.4 years /25/. All participants had been diabetic for at least 10 years and were divided into two treatment groups: one undergoing intensive treatment with an HbA1c goal < 6.0% and the other undergoing regular treatment with a goal of 7–7.9%. The risk of mortality increased continuously with HbA1c between 6.0% and 9.0%, but was the highest in the group undergoing intensive treatment with values above 7.0%.

Prognostic relevance

The prognostic relevance of HbA1c for the assessment of an increased risk of diabetes was summarized in a review of 16 prospectives studies /26/. The results demonstrated:

  • The annualized incidence of diabetes ranges from 0.1% at HbA1c below 5.0% to 54.1% at above 6.1%
  • Above the HbA1c range of 5.5–6.5%, the incidence of diabetes increases steeply
  • HbA1c values between 6.0% and 6.5% are associated with a greatly increased risk of diabetes (incidence of 25–50% in 5 years) compared to values below 5%)
  • HbA1c between 5.5% and 6.0% are associated with a moderately increased risk of diabetes (incidence of 9–25% in 5 years) compared to values below 5%
  • HbA1c between 5.0 and 5.5% is associated with a slightly increased risk of diabetes (incidence below 9% in 5 years) compared to below 5%.

Table 3.6-3 Analytical goals for an HbA1c method with reference to the IFCC reference method /17/




Total error


≤ 0.6

≤ ± 0.9

≤ ± 2.0


≤ 1.3

≤ ± 1.9

≤ ± 3.9


≤ 1.9

≤ ± 2.8

≤ ± 5.9

Table 3.7-1 Functional tests for the diagnosis and differentiation of hypoglycemia syndrome

Functional test

Clinical and laboratory findings

72-h fast /8/

Principle: glucose homeostasis is maintained by the fine balance between the hormones insulin and glucagon. An impairment of this hormonal balance (e.g., due to autonomous production of insulin), leads to hyperinsulinism, and under fasting conditions hypoglycemia will develop. In the 72-h fast, hypoglycemia is provoked, and blood samples taken in the hypoglycemic state are tested at the laboratory to determine the cause of the hypoglycemia.

Test procedure: the patient should not eat for up to 72 h.

  • The fast starts at the time of the last food intake
  • Medications should be discontinued for 3 days prior to testing, if possible
  • Calorie- and caffeine-free drinks are allowed
  • During waking hours the patient should move
  • Samples for glucose are obtained every 2 h unless the glucose is ≤ 60 mg/dL (3.3 mmol/L), in which case samples should be obtained at least hourly. In addition 3 mL of blood should be collected to determine C-peptide, insulin and, if necessary, proinsulin from the same sample
  • Stop the fast when the glucose concentration falls to ≤ 45 mg/dL (2.5 mmol/L) in venous plasma and capillary plasma or to < 40 mg/dL (2.2 mmol/L) in venous and capillary whole blood and the patient notices symptoms of hypoglycemia
  • A the end of fasting, determine glucose, insulin, C-peptide and, if necessary, proinsulin and β-hydroxy butyrate from the same sample. Then administer 1 mg of glucagon intravenously and measure plasma glucose after 10, 20 and 30 min.
  • If cortisol, growth hormone or glucagon deficiency is suspected, determine these hormones at the beginning and at the end of the fast.

In the case of children, the duration of fasting depends on the age of the child. It should be no longer than 12 h for children aged < 1 year and no longer than 24 h for older children. In children, the 72-h fast is mainly used to provoke congenital metabolic disorders which can cause hypoglycemia, such as fatty acid oxidation disorders (see Section 3.2 – Hypoglycemia syndromes).

test /7/

Principle: the secretion of the β cells, determined by measuring C-peptide levels in plasma, is monitored under insulin-provoked hypoglycemia. In healthy individuals, C-peptide secretion decreases, with a decline in plasma C-peptide levels. In patients with autonomous insulin and C-peptide secretion, the decline in C-peptide plasma level is small.

Test procedure: the C-peptide suppression test should be performed only if the glucose concentration prior to the test is ≥ 60 mg/dL (3.3 mmol/L). An indwelling intravenous catheter is placed in a suitable arm vein. Regular insulin (0.12 U/kg/h) and body weight is administered as an intravenous infusion over a period of 60 min. Samples for glucose and C-peptide are collected prior to the test and at 10 min. intervals.

test /9/

Principle: following administration of tolbutamide, patients with autonomous insulin secretion exhibit a different secretion pattern for insulin and C-peptide than healthy individuals.

Test procedure: the test should be performed only if the glucose concentration prior to the test is ≥ 60 mg/dL (3.3 mmol/L). The patient is administered an intravenous dose of 1 g of tolbutamide (5% aqueous solution) over a 3-min. period; children are administered 25 mg/kg body weight, but no more than 1 g in total. Blood samples of 3 mL each are collected prior to the test as well as 5, 10, 20, 30, 40, 60, 90, 120 and 180 min. after the test to determine glucose, insulin and C-peptide.

test /9/

Principle: an intravenous bolus injection of glucagon leads to maximum counter regulatory secretion of insulin and C-peptide. However, in healthy individuals the plasma concentrations of both hormones do not reach the concentration that is measured in patients with an islet cell adenoma.

Test procedure: the patient should eat carbohydrate-rich foods for at least 3 days and then fast for 8 hours prior to the test, if possible. 1 mg of glucagon, diluted in 10 mL of physiological NaCl, is gradually injected as an intravenous bolus. Then blood samples of 3 mL each are taken prior to the test and after 1, 5, 10, 15 and 30 min. to determine glucose, insulin, C-peptide and, if necessary, β-hydroxybutyrate.

Table 3.7-2 Insulin, C-peptide and proinsulin reference intervals

Insulin /4/

Postabsorptive phase (> 6 h after the last food intake)

2–23 mU/L (14–160 pmol/L)

Prolonged fasting* (12 h and longer, incl. 72-h fast)

< 6 mU/L (42 pmol/L)

Within 30 min. in oGTT or after glucagon stimulation

50–200 mU/L (347–1389 pmol/L)

C-peptide /5/

Postabsorptive phase

1.0–2.1 μg/L (0.3–0.7 nmol/L)

Prolonged fasting (72-h fast)*

< 0.7 μg/L (0.2 nmol/L)

90 min. after a meal (600 kCal) /5/

3.6–40 μg/L (0.5–5.5 nmol/L)

Proinsulin /6/

Postabsorptive phase

17–103 ng/L (1.8–11 pmol/L)

Within 30 min. in oGTT

63–848 ng/L (6.7–90.3 pmol/L)

* Prolonged fasting with blood glucose levels < 60 mg/dL (3.3 mmol/L); conversion to molarities:

– Insulin: mU/L × 6.945 = pmol/L

– C-peptide: μg/L × 0.331 = nmol/L

– Proinsulin: ng/L × 0.106 = pmol/L

Values are the 2.5th and 97.5 percentiles

Table 3.7-3 Clinical significance of functional tests for hyperinsulinism-induced hypoglycemia

Functional test

Clinical and laboratory findings

72-h fast

An insulinoma can be excluded if venous and capillary plasma glucose does not fall to below 45 mg/dL (2.5 mmol/L) and venous and capillary whole blood glucose does not fall to below 40 mg/dL (2.2 mmol/L) within 72 hours. Concomitant insulin levels are < 6 mU/L (42 pmol/L) and C-peptide levels are < 0.7 μg/L (0.2 nmol/L).

Insulin levels ≥ 6 mU/L (42 pmol/L) and C-peptide levels ≥ 0.7 μg/L (0.2 nmol/L) in the presence of hypoglycemia are indicative of an insulinoma. No excessively elevated insulin and C-peptide concentrations are seen in insulinoma. A study /9/ reported insulin levels in the range of 6–70 mU/L (42–490 pmol/L) in the majority of cases, and levels < 6 mU/L (42 pmol/L) in 10% of insulinoma patients with hypoglycemia. Hypoglycemia occurred as follows in the insulinoma patients participating in the study: in 29% of patients within 12 h after the last meal, in 72% within 24 h, in 92% within 48 h and in 98% within 72 h.

C-peptide suppression test

The test can be used as a primary investigation if an insulinoma is suspected, or when the 72-h fast is not conclusive. The relative decline in C-peptide concentrations after 60 min. of insulin infusion is determined and related to the lean body mass. In a study /10/ of healthy individuals, the decline varied between 67% in lean, young individuals and 71% in obese, older individuals. A smaller decline in C-peptide secretion indicates an insulinoma.

Tolbutamide test

The test can be used as a primary investigation if an insulinoma is suspected, in particular in obese patients. The presence of an insulinoma is indicated if mean venous plasma glucose, determined from the 120, 150 and 180 min.-samples, is ≤ 56 mg/dl (3.1 mmol/l) in slender patients and ≤ 61 mg/dl (3.4 mmol/l) in obese patients. The diagnostic sensitivity of the test for diagnosing an insulinoma is reported to be 95%, with > 95% specificity /9/.

Glucagon test

Using the criterion of a peak insulin level > 130 mU/L (910 pmol/L) or an increase in insulin > 100 mU/L (700 pmol/L) after overnight fasting as compared to baseline, the diagnostic sensitivity of the glucagon test for diagnosing an insulinoma is 50–80% /9/. The maximum increase in C-peptide is > 2.5 μg/L (0.7 nmol/L). Hydrochlorothiazide, diphenylhydantoin and diazoxide can cause false negative, and tolbutamide, aminophylline and obesity can cause false positive results. Approximately 80% of insulinoma patients have proinsulin levels that make up > 20% of the insulin value. The maximum increases in insulin, C-peptide and proinsulin can be assessed from Fig. 3.7-1 – Glucose, insulin, C-peptide, proinsulin and β-hydroxy butyrate at the end of the 72-h fast.

Table 3.7-4 Interpretation of the 72-h fast*, with kind permission from Ref. /8/


Hypoglycemia symptoms

Glucose (mg/dL)

Insulin (mU/L)

C-peptide (nmol/L)

Proinsulin (pmol/L)

β-hydroxy-butyrate (mmol/L)

Glucose change (mg/dL)

Sulfonylurea in plasma


> 45

< 6

< 0.2

< 5

> 2.7

< 25



≤ 45

≥ 6

≥ 0.2

≥ 5

≥ 2.7

≥ 25

Factitious hypoglycemia

  • Insulin


≤ 45

≥ 6

< 0.2

< 5

≤ 2.7

≥ 25

  • Sulfonyl­urea


≤ 45

≥ 6

≥ 0.2

≥ 5

≤ 2.7

≥ 25


Postprandial reactive hypoglycemia


≤ 45

≥ 6

< 0.2

< 5

≥ 2.7

< 25

Non-insulin-related hypoglycemia


≤ 45

< 6

< 0.2

< 5

> 2.7

< 25

Food intake during the test

> 45

< 6

< 0.2

< 5

≤ 2.7

≥ 25

No hypoglycemic disease


> 45

< 6

< 0.2

< 5

> 2.7

< 25

* Measurements at the end of the fasting test

Table 3.7-5 Diseases and conditions associated with hyperinsulinism and hypoglycemia

Clinical and laboratory findings


The incidence of insulinoma is 4 cases per 1 million population per year. Insulinomas account for approximately 60% of islet cell tumors are more common in adults than in children. They are hyper vascular, small single tumors, generally of 2 cm diameter. Approximately 10% of insulinomas are malignant and have metastases, and 10% are multiple, in particular in patients with multiple endocrine type 1 neoplasia (MEN 1). Approximately 50% of patients with multiple insulinomas have MEN 1. Approximately 5% of insulinomas are associated with MEN 1, and 21% of patients with MEN 1 will develop an insulinoma /11/.

MEN 1 must be diagnosed by excluding:

  • Hyperprolactinemia caused by a prolactinoma
  • Hyperparathyroidism due to parathyroid hyperplasia
  • Hypergastrinemia due to a gastrinoma.

Insulinoma patients frequently present due to sudden episodes of confusion and impaired consciousness which often started years ago and have recently become more frequent. Episodes typically occur more than 6 h after meals or after overnight fasting, before breakfast or after physical work.

Laboratory findings: the diagnosis of organic hyperinsulinism is confirmed when plasma glucose is ≤ 40 mg/dL (2.2 mmol/L), insulin is ≥ 6 mU/L (42 pmol/L) and the C-peptide level is ≥ 230 pmol/L in the presence of clinical symptoms. Some investigators set different thresholds: glucose less than 45 mg/dL (2.5 mmol/L), insulin ≥ 50 pmoL/L and C-peptide ≥ 300 pmoL/L). If these criteria are not met, the 72-h fast is the primary diagnostic assay. Approximately 98% of insulinomas are detected through a 72-h fast. In approximately 90% of patients the insulin and C-peptide levels are higher than the fasting range of healthy individuals. Up to 95% of insulinoma patients have elevated proinsulin levels, which account for at least 25% of the total immunoreactive insulin and are usually in the range of 180–2700 ng/L (20–300 pmoL/L) /12/. Other diagnostic tests see Tab. 3.7-1 – Functional tests for the diagnosis and differentiation of hypoglycemia syndrome.

Postprandial hyperreactive hypoglycemia (PRH)

Some patients with PRH have functional hyperinsulinism. If glucose levels are < 45 mg/dL (2.5 mmol/L) they can be associated with a pronounced increase in insulin. This occurs either in patients who are overweight and have impaired glucose tolerance, or in cases of isolated PRH. The release of insulin is often slow or delayed with reference to peak glucose levels /13/. Also refer to Section 3.2 – Hypoglycemia syndromes.

Factitious hypoglycemia

Nondiabetic individuals who inject insulin or take oral antidiabetic drugs include in particular women who are in their third and fourth decade of life and often work in a medical profession. If insulin was administered frequently, insulin antibodies may be detectable. Insulin, C-peptide and proinsulin levels are elevated in individuals who take oral antidiabetic drugs /9/. A problem of uncertain magnitude arises from medication error such as the administration of insulin or sufonylurea to the patient instead of the diabetic patient in the next bed.

Factitious hypoglycemia can be differentiated from insulinoma only by determining the plasma sulfonylurea concentration. If the hypoglycemia is induced by injecting insulin, this can be detected by the determination of insulin and C-peptide. The insulin level can be > 100 mU/L (700 pmol/L), C-peptide concentrations are suppressed to the limit of detection.

Autoimmune insulin hypoglycemia

These patients have spontaneous insulin antibodies even if they have never received insulin /9/. The hypoglycemia can be severe in these patients and can occur during fasting or after meals. While the disease can manifest in neonates, it is also possible that clinical symptoms may not appear until later in life. As with hypoglycemia factitia, C-peptide is suppressed but insulin levels are raised.

Neonatal hyperinsulinism

Hyperinsulinism occurs in macrosomic neonates and children born to mothers with diabetes or gestational diabetes mellitus or to mothers who were administered glucose or medications. A study /14/ found that macrosomic neonates had significantly higher cord blood insulin levels (on average 19 mU/L (137 pmol/L)) compared to normal-weight term neonates (8.7 mU/L; 63 pmol/L). These neonates can easily develop hypoglycemia.

Congenital hyperinsulinemia /15/

Congenital hyperinsulinemia typically presents as persistent hypoglycemia in newborns. It is a heterogeneous disease in terms of clinical manifestation, genetic etiology and treatment response. At present, five gene mutations are known. In most cases the condition is caused by mutation of the genes that encode one of the two protein units of the ATP dependent β-cell potassium channel (ABCC8 and KCNJ11).

Laboratory findings: hyperinsulinemia, hypoglycemia, ketone bodies (β-hydroxybutyrate) reduced, decrease in free fatty acids.

Table 3.7-6 Interpretation of baseline and glucagon-stimulated C-peptide levels in diabetics /18/


5-, 10-min. level


< 1.0 (0.33)

< 1.5 (0.50)

No insulin reserve

≥ 1.0 (0.33)

≥ 1.5 (0.50)

Insulin reserve positive

Data expressed in μg/L (nmol/L)

Table 3.7-7 Tests for the assessment of β-cell function and insulin resistance

Clinical and laboratory findings

Assessment of first-phase insulin response

Principle: the initial (first-phase) insulin response occurs within 30 seconds of intravenous glucose administration and reaches a peak after 3 to 5 minutes with a secretion rate of 80 pg per minute and β-cell islet. After 10 min., insulin levels are back to baseline. During the first-phase response, preformed insulin is released from the secretory granules of the islet cells. The second, delayed release after 10 min. lasts as long as glucose remains elevated; during this phase, stored insulin, newly produced insulin and proinsulin are secreted. The ability of the endocrine pancreas to respond adequately to a glucose stimulus, both in terms of quantity and on a moment to moment basis, characterizes the endocrine plasticity and describes the capacity of the islet cells to maintain glucose homeostasis.

Procedure: 0.5 g of glucose/kg body weight and a maximum of 35 g are administered as an intravenous bolus in the form of a 25% glucose solution within a period of 3 min. To determine glucose, insulin and, if necessary, C-peptide, blood samples of 3 mL each are drawn at the end of the glucose infusion (0 value) and 1, 3, 5 and 10 min. post infusion.

Assessment: a faulty insulin secretion characteristic, in particular the loss of the first-phase insulin response, precedes other manifestations of type 2 diabetes. This has a significant effect on glucose metabolism, since hepatic glucose production is not immediately suppressed, resulting in postprandial hyperglycemia. In addition, significant postprandial changes in glucose levels can promote insulin resistance.

Clamp test /20/

Glucose clamp testing is the gold standard for the diagnosis of insulin resistance and insulin secretion of the β-cells. There are two methods: the euglycemic clamp and the hyperglycemic clamp. Both are predominantly used for research purposes.

Principle of the hyperinsulinemic euglycemic clamp: insulin is infused at a constant rate. Glucose is infused via a second pump so that a glucose concentration of approximately 90 mg/dL (5 mmol/L) is maintained. The glucose rate required to maintain a constant glucose concentration is measured. The euglycemic clamp quantifies the insulin sensitivity by directly registering the effect of insulin on glucose consumption under steady-state conditions (insulin resistance = 1/insulin sensitivity). The degree of insulin resistance is inversely proportional to the level of the glucose infusion rate. According to a study /21/, the tissue glucose uptake in non-insulin-resistant, insulin-resistant and type 2 diabetics is 46.0 ± 16.9, 19.1 ± 4.6 and 17.1 ± 8.2 μmoL/min. per kg of lean body mass respectively (mean value and standard deviation).

Principle of the hyperinsulinemic hyperglycemic clamp: insulin is infused at a constant rate. Glucose is infused via a second pump at variable rates so that a glucose concentration of approx. 216 mg/dL (12 mmol/L) is maintained. Once the glucose concentration is stable, the glucose infusion rate is measured and divided by the insulin infusion rate. This ratio is an indicator of β-cell secretion.

Assessment: patients with type 1 diabetes have no insulin secretion and no insulin resistance (normal insulin sensitivity). Insulin sensitivity is defined as = 1/insulin resistance. Patients with early-stage type 2 diabetes have high insulin resistance and high insulin secretion (obese individuals in particular have high insulin resistance). In the late stage of type 2 diabetes, insulin secretion is reduced. Fig. 3.7-2 – Insulin sensitivity and glucose uptake index for adolescents with type 2 diabetes and overweight adolescents shows the insulin resistance in the form of insulin sensitivity; Fig. 3.7-3 – Time curve of insulin concentrations during the hyperglycemic clamp in adolescents with type 2 diabetes and overweight adolescents shows the time curve of insulin secretion during the hyperglycemic clamp.

Homeostasis Model Assessment (HOMA)

In the fasting state the insulin level is dependent on the glucose concentration. Healthy individuals and type 2 diabetics have individual glucose-insulin ratios which only depend on the time of the last food intake. If fasting hyperglycemia is present, the extent is an interaction of insulin resistance and β-cell function. The HOMA is a method to assess insulin resistance and β-cell function and requires only fasting glucose and insulin levels (Fig. 3.7-4 – Homeostasis Model Assessment (HOMA)). A distinction is made between:

  • The Homeostasis Model Assessment of Insulin Resistance (HOMA-IR), the results are directly related to insulin resistance
  • The Homeostasis Model Assessment of β-cell function (HOMA-β cell) for the assessment of the fasting insulin secretion /23/.


The fasting plasma glucose (FPG) and fasting plasma insulin (FPI) levels are determined, and the HOMA-IR is calculated as follows: HOMA-IR = FPG (mmol/L) × FPI (mU/L)/22.5. The constant of 22.5 was introduced under the assumption that young, healthy individuals have an insulin resistance of 1.0. The HOMA-IR correlates with the hyperinsulinic glucose clamp and is a predictor of impaired glucose tolerance (prediabetes) and type 2 diabetes.

In individuals with insulin resistance the HOMA-IR is related to the BMI /21/:

  • The HOMA-IR is > 4.65 (FPI ≥ 20.7 mU/L) if the BMI is > 28.9 kg/m2
  • The HOMA-IR is > 3.60 (FPI ≥ 16.3 mU/L) if the BMI > 27.5 kg/m2

HOMA-β cell

Calculation: fasting plasma glucose (FPG) and fasting plasma insulin (FPI) levels are determined and calculated as follows: HOMA-β cell = 20 × FPI (mU/L)/[FPG (mmol/L) – 3.5].

Healthy individuals aged < 35 years who have 100% β-cell function have a HOMA-β cell of ≥ 100, obese individuals a value ≥ 200 and long-term type 2 diabetics have levels < 100.

The HOMA has its limitations /24/. For example, the HOMA-IR is not a good predictor of insulin resistance in older individuals with impaired glucose tolerance and type 2 diabetes. Depending on the calculator, a factor of 6 or 7 is used for converting insulin values from mU/L to pmol/L. Another important factor is the insulin test used, because results obtained with different assays may vary by up to 50%.

Quantitative Insulin Sensitivity Check Index (QUICKI) /2526/

The QUICKI index is a test for assessing insulin sensitivity which shows results comparable to those obtained with the hyperinsulinemic-euglycemic clamp. The following equation is used for the calculation /25/: QUICKI index = 1/[log (I0)+ log (G0)].

G0 ,fasting glucose in mmol/L; I0 , fasting insulin in mU/L. The QUICKI reference intervals (x ± s) are:

  • 0.382 ± 0.007 in normal-weight individuals
  • 0.331 ± 0.010 in obese individuals
  • 0.304 ± 0.007 in type 2 diabetics.

Figure 3.1-1 Development of type 1A diabetes. With kind permission from Marker J, Maclaren N. Clin Lab Med 2001; 21: 15–30. Abbreviations: CTLA-4, cytotoxic T-lymphocyte adhesion ligand polymorphism on chromosome 2q; FPIR, first-phase insulin response; GAD, glutamic acid decarboxylase; Glima-38A, 38 kDa islet cell antigen; IA, islet cell antigens.

9 months – 3 years Months to years Months Island cellsmass IL-12/IFN-γ Genetic predisposition HLA-DR/DQ ? A.B.C CTL A-4 other Lack of protection by NK-cells and CD 25+ -T-cells Island cellsantigensIAA, GAD65A, IA-2A, IA-2βA, glima-38A Metabolic dysfunctions FPIR im ivGTT Inductive result, e.g. Coxsackievirus infection Cell-mediatedresponse onisland antigens Start of the diabetes Honeymoon phase Disturbed oGTT tolerance Increased fasting glucose

Figure 3.1-2 Relationship between maximum insulin response to glucose and insulin sensitivity in normoglycemic individuals (NGP), relatives of type 2 diabetics (VType 2 DM), type 2 diabetics (type 2 DM), older individuals, former gestational diabetes (FGDM), impaired glucose tolerance (IGT), and women with polycystic ovary syndrome (PCOS). In each case, the percentiles 5, 25, 50 and 75 are stated. The tests were performed using the clamp technique (see Tab. 3.7-7 – Tests for the assessment of β-cell function and insulin resistance). With kind permission from Ref. /13/.

PCOS IGT Insulin response (pmol/L)700600500400300100100 0 0 1 2 3 4 5 6 7 20 FGDM Older People Type 2-DM VTyp 2-DM NGP 75. 50. 25. 5. Index of insulin sensitivity s i (× 10 –5 min –1 /pmol/L)

Figure 3.1-3 Relationship between maximum insulin response to arginine stimulation and the plasma glucose concentration in healthy controls () and type 2 diabetics (). Modified from Ref. /15/.

2,000 1,500 1,000 1.500 0 100 200 300 400 500 600 Glucose (mg/dL) Insulin (pmol/L)

Figure 3.1-4 Development of type 2 diabetes in three stages. Type 2 is the final stage of years or decades of prediabetes.

Plasma glucose Glykogenolysis Gluconeogenesis Cell. glucose transport Inappropriate Insulin secretion Athero- genesisHyper-insulin-emiaInsulinresistence TG HDL Arterialhyper-tension Diabetes genes Micro- angio- pathy Macro- angio- pathy Stage 3 Typ 2 diabetes Stage 2 Fasting glucosePrediabetes Stage 1 Normalglucose tolerance Lipo-genesisadiposity Hip circumferenceBMI

Figure 3.1-5 Important organs and tissues involved in glucose metabolism in the postabsorptive state (top) and postprandial state (bottom). During the transition from the postabsorptive to the postprandial state, the primary site of glucose uptake shifts from insulin-independent tissues (e.g., the brain) to insulin-dependent tissues (e.g., the liver, muscle, and adipocytes). FFA, free fatty acids. Modified with kind permission from Ref. /18/.

Brain Liver Stomach/Intestine Muscle Lactate Glycogen CO 2 indirectly Glycogen Glycogenolysis directly Gluconeogenesis GlycerolLactate Liver Brain Muscle Gluconeogenesis Adipose tissue Adipose tissue GlycerolLactate AlanineGlutamine Glykogenlysis FFA FFA

Figure 3.1-6 Differentiation of MODY from type 1 and type 2 diabetes. With kind permission from Fehrmann HC, et al. Dtsch Ärztebl 2004; 101: B719–25.

Diabetes mellitus Overweight No Yes Autoimmune phenomena Autosomal-dominant heredity in > 3 generations Postive Negative Positive Negative MODY Type 2 Type 1

Figure 3.2-1 Activation of insulin counter regulatory hormones and clinical symptoms as a function of decreasing blood glucose levels in healthy individuals and diabetics. With kind permission from Ref. /7/.

Blood glucose(mmol/L) (mg/dL) 4        (72) 3        (54) 2        (36) 1        (18) Adrenalinerelease Start of brain dysfunction Confusion/loss ofconcentration Coma/cramps Permanent brain damage Sweating,tremor

Figure 3.2-2 Flow chart for the differentiation of hypoglycemia in adults, modified from Ref. /8/.

Drugs– Insulin– Oral hypoglycemic drugs– Salicylates– Quinine– Alcohol History Suspected Hypoglycemia Glucose < 60 mg/dL (3.3 mmol/L) Fasting hypoglycemic symptoms(spontaneous onset, after 72-hour fast) Postprandial hypoglycemicsymptoms Rule out fasting C-peptide, insulin Glucose < 60 mg/dL (3.3 mmol/L)after a standard meal Gastrointestinal surgeryMotility disturbance No Gastrointestinal surgery– Liver disease– Postprandial reactive hypoglycemia Increased Normal Rule out factitious hypoglycemia Insulinoma Consider– Uremia– Sepsis– Tumor– Autoimmune disease

Figure 3.2-3 Hypoglycemia-associated autonomic failure in insulin-dependent diabetics. With kind permission from Ref. /32/.

Absolute insulin deficiency Imperfectinsulin adjustment Reduced autonomic (including adrenal medulla) answers No hyperglycemia No reduced insulin supply No hyperglycemie perception Defective glucosecounterregulation No glucagon increase Barely symptoms Barely adrenalin answer Type 1-diabetes

Figure 3.2-4 Galactose metabolism. Enzyme-catalyzed reactions: 1, Galactokinase; 2, Galactose-1-phosphate uridyltransferase; 3, UDP-galactose-4-epimerase; 4, UDP-glucose-phosphorylase; 5, Aldose reductase; 6, Galactose dehydrogenase.

Galacit Galactose Galactose-1 phosphate UDP galactose UDP glucose UDP glucose Glucose-1phosphate Lactose Galactonic acid 5 6 6 4 2 3 5 1

Figure 3.3-1 Conversion factors for glucose, measured with 2 different methods and specimens. With kind permission from Ref. /29/. E, conversion recommended; NE, conversion not recommended, but used for rough estimate.

Biosensor-glucose Plasma glucose Whole blood glucose (NE) 1.11 (NE) 1.18 (E) 0.94

Figure 3.4-1 Urinary glucose excretion (ordinates) as a function of blood glucose concentration (abscissae). With kind permission from Ref. /6/.

mg/min.250 200 150 100 50 0 mmol/ min.13.9 11.1 8.3 5.6 2.8 80 100 120 140 160 4.4 5.6 6.7 7.8 8.9 mmol/L mg/dL

Figure 3.5-1 Prevalence of newly detected type 2 diabetes (T2D), impaired glucose tolerance (IGT) and impaired fasting glucose (IFG) and in individuals with normal oGTT (NGT) according to the old World Health Organization (WHO) criteria and the new American Diabetes Association (ADA) criteria. With kind permission from Ref. /19/.

Old WHO criteria: an oGTT was performed in patients with glucose levels of 110–200 mg/dL (6.1–11.1 mmol/L). Criteria for the diagnosis of IGT were a 2-h value > 140 mg/dL (7.8 mmol/L) and for T2D a 2-h value > 200 mg/dL (11.1 mmol/L).

ADA criteria: an oGTT was performed in patients with glucose levels of 100–125 mg/dL (5.6–6.9 mmol/L). The test criteria were the 120-min glucose values in Tab. 3.5-1 – Criteria for diagnosis of diabetes mellitus and impaired glucose tolerance using 75 g oGTT. According to these criteria, more patients have T2D.

706050403020100 Prevalence (%) Old CriteriaNew Criteria IFG Without IGT IGT Typ 2Diabetesmellitus NGT

Figure 3.6-3 Glycation of N-terminal valine of hemoglobin with glucose and subsequent Amadori rearrangement. K+1 = 0,9 × 10–3 mmol × h–1, K–1 = 0.35 h–1, K2 = 0.0055 h–1.

Hb-NH 2 + O CH H OH H HO H OH H OH CH 2 OH Hb N CH H C OH HO C H H C OH H C OH CH 2 OH Glucose Aldimine (Schiff base) Ketoamine K +1 K –1 K 2 Hb N CH 2 C O C C C H HO H OH H OH CH 2 OH H C C C C

Figure 3.6-1 Flow chart for the diagnosis of diabetes based on the determination of HbA1c, fasting plasma glucose (FPG), and the oral glucose tolerance test (oGTT) /4/. Plasma glucose levels are expressed in mmol/L. * If diabetic symptoms are present, a glucose determination is performed in addition. ** In suspicion of a false HbA1c result (see “Comments and problems”), the diagnosis is based on glucose levels.

Diabetes mellitus symptomatology (weight loss, polyuria, polydipsia) and or increased diabetes riskLaboratory diagnostics HbA 1c */** 5.7–6.4%39–47 mmol/mol FPG ≥ 5.6–6.9and/or 2h-oGTT-PG7.8–11.0 FPG < 5.6and/or FPG < 5.6 und 2h-PG ≥ 11.1 FPG ≥ 7.0and/or 2h-oGTT-PG≥ 11.1 Fasting plasma glucose (FPG) oder oGTT Clarification of the dia- betes risk, lifestyle intervention, therapy of risk factors. Renewed risk investigation HbA 1c after 1 year. < 5.7%< 39 mmol/mol ≥ 6.5%≥ 48 mmol/mol No diabetesmellitus Diabetesmellitus Therapy according guidelines

Figure 3.6-2 The endoproteinase Glu-C cleaves the N-terminal 6 amino acids of the hemoglobin β-chain from HbA0 and HbA1c leaving the C-terminal side of glutamine (Glu). HbA0 and HbA1c are determined quantitatively and the proportion (percentage and molar ratio) of HbA1c to HbA0 is formed /5/. Common hemoglobin variants are listed.

Val His Leu Thr Pro Glu Val His Leu Thr Pro Glu N-terminal of the hemoglobin A β-chain Hemoglobin variants: HbS: In position 6 glutamine is replaced by valine HbC: In position 6 glutamine is replaced by lysine HbD: Amino acid variations in different position (HbD clinical unobtrusively) HbE: In position 26 glutamine is replaced by lysine (clinical conspicuous especially in combination with a thalassemia) HbA 0 -peptie EndoproteinaseGlu-C 6 1 146 HbA 1c -peptide

Figure 3.7-1 Glucose, glucose change, insulin, C-peptide, proinsulin and β-hydroxybutyrate at the end of the 72-h fast or in the glucagon test in 25 normal individuals and 40 patients with an insulinoma. Bright regions represent glucose levels < 50 mg/dL (2,8 mmol/L). The vertical black lines represent the criteria for diagnosing an insulinoma, i.e., insulin ≥ 6 mU/L (42 pmol/L), C-peptide ≥ 0,7 μg/L (0.2 nmol/L), proinsulin ≥ 45 ng/L (5 pmol/L), β-hydroxybutyrate ≤ 2.7 mmol/L and glucose change ≥ 25 mg/dL (1.2 mmol/L). With kind permission from Ref. /8/.

90 80 70 60 50 40 30 20 Normal 22 5 (mg/dL) 18 36 Insulinoma 0.1 1.0 2.7 β -hydroxybutyrate (mmol/L) ( mg/dL) Plasma glucose 1.0 2.0 5. 0 Change in glucose (mmol/L) 5.00 4.44 3.89 3.33 2.78 2.22 1.67 1.11 (m m ol/L) 90 80 70 60 50 40 30 20 10.0 Normal Insu- linom a 90 Normal Insulinoma 15 10 20 100 800 Proinsulin (pmol/L ) Normal Insulinoma 0.03 0.1 0.5 C- peptide (nmol/L) 0.2 1.0 (m m ol/L Normal Insulinoma 56 14 70 ( m U/L) 140 36 100 500 1,000 (pmol/L) Insulin 5.00 4.44 3.89 3.33 2.78 2.22 1.67 1.11 Plasma glucose (mg/dL)

Figure 3.7-2 Insulin sensitivity (1/insulin resistance) and glucose uptake index for adolescents with type 2 diabetes and overweight adolescents. Insulin sensitivity is low (insulin resistance high) and the glucose uptake index is low in type 2 diabetes (see bar on the right). With kind permission from Ref. /23/.

800 0 1,600 2,400 3,200 0 1 2 3 Insulin sensitivity(μmol/kg/min per pmol/L) Glucose disposition index(mmol/kg/min.)

Figure 3.7-3 Time curve of insulin concentrations during the hyperglycemic clamp in adolescents with type 2 diabetes (T2DM) and in overweight adolescents (OBC). With kind permission from Ref. /22/.

T2DM OBC –30 –15 0 0 2,000 1,600 1,200 800 400 15 30 45 60 75 90 105 120 Time (min) Insulin (pmol/L)

Figure 3.7-4 Homeostasis Model Assessment (HOMA). Prediction of insulin resistance and β-cell function based on fasting glucose and insulin concentration. The grid shows, as a function of glucose and insulin, the insulin resistance in stages R-1 (normal) to R-16 (high resistance) and the β-cell function from 200% (β-200) to 12.5% (β-12.5). Modified with kind permission from Ref. /23/.

R-16 R-8 R-4 R-2 R-1 R-0,5 α = 200% β = 100% β = 50% β = 25% β = 12,5% R-8 R-16 R-4 R-2 R-1 R-0,5 0 0 5 10 15 20 25 30 35 40 1 2 3 4 5 6 7 8 9 10 11 12 13 Decreasing β- cell function (β) Increasing insulin resistance (R) Fasting plasma glucose (mmol/l) Fasting insulin (mU/l)

Figure 3.7-5 Biosynthetic pathway from preproinsulin to insulin in the β-cells of the pancreas. Preproinsulin is synthesized in the rough endoplasmic reticulum (RER) and converted to proinsulin by cleavage of the signal peptide and folding. It is then transported to the Golgi apparatus (GOLGI) where it is stored in immature secretory granules. During the granule maturation process, proinsulin is converted to insulin and C-peptide by the concerted action of calcium ions, proprotein convertases and carboxypeptidase E which are present in an acidic environment /32/.

10–15 min. 30–40 min. 20–40 min. 2–4 h 1–2 days Preproinsulin synthesis Secretory pathway(regulated secretion)– Insulin + C-peptide about 96%– Proinsulin + intermediates about 4% Constitutive pathway(unregulated secretion)– Proinsulin about 1–2%– Membrane proteins Proinsulin Energy dependent transfers Prohormone and converting enzym Clathrin coat Insulin crystalC-peptide Mature granules Plasma membrane RER Trans Lysosome Golgi

Figure 3.7-6 Processing of proinsulin to insulin and C-peptide. With kind permission from Ref. /31/. The right branch is the dominant pathway. Proprotein convertase 3 (PC3) generates the des-31,32 proinsulin intermediate, which is a preferred substrate of proprotein convertase 2 (PC2). The left branch is more active than the right one under conditions where PC3 activity is low and/or there is a lack of PC2. Following the action of the PCs, carboxypeptidase E (CPE) catalyzes the cleavage of Lys-Arg or Arg-Arg residues from the newly formed C-terminal ends.

C-peptide A-chain B-chain Proinsulin