Glucose Tolerance Tests

History and Definition of Oral Glucose Tolerance Tests

A major breakthrough in the understanding of glucose intolerance as a basis and risk factor for development of type 2 diabetes and cardiovascular diseases was the introduction of a worldwide standardization of the OGTT in the 1970s. By this introduction, glucose tolerance became a standardized entity, which enabled studies in metabolism, physiology, and clinical medicine with detection of risk factors as well as progressive follow-up studies using a standard recognized worldwide. At the same time, and also of significant importance for the generation of present-day knowledge within the field, was the introduction of the clinical entity impaired glucose tolerance (IGT), which replaced the term ''borderline'' diabetes. A problem with the term borderline diabetes was that its definition was not uniform, which was partly due to inconsistencies in the procedure of performing a glucose tolerance test, with the amount of glucose ingested varying from 50 to 100 g or given on a kilogram basis. IGT as an entity was thus introduced simultaneously with the suggestion that glucose tolerance in a clinical test should be determined following ingestion of 75 g glucose, with a blood sample for the measure of glucose to be taken 2 h later.

The evaluation of the standardized OGTT in the clinical setting relies on the 2-h glucose value. This value during the 75 g OGTT usually displays a normal distribution slightly skewed to the right. Figure 1 shows the distribution pattern of 2-h glucose levels obtained from 802 Caucasian subjects in the Malmo Prevention Study, an epidemiological study from Sweden. From this distribution, normal values may be defined statistically from mean and variance values for statistical definition of the distribution. The mean value, as in most studies, is «7mmol/l and standard deviation is «1mmol/l. By defining reference values as 95% confidence intervals, the cutoff value for normality would be

2-h Glucose (mmol/l)

Figure 1 Distribution of the 2-h glucose value in an OGTT performed in 802 Caucasian subjects from the Malmo Prevention Study (unpublished data).

approximately 9 mmol/l and, hence, values higher than 9 mmol/l would indicate diabetes. By using such a definition of diabetes, a large number of subjects would have the disease, the clinical relevance of which is doubtful. Therefore, the definition of diabetes has instead been based on prospective studies evaluating the risk for microvascular disease and the cutoff-levels have been defined as levels increasing this risk. Therefore, a cutoff value of 11.1 mmol/l glucose has been used for the definition of type 2 diabetes.

OGTT was frequently used during the 1980s and 1990s for the clinical diagnosis of type 2 diabetes and in epidemiological studies, which markedly increased our knowledge of these conditions. By the end of the 1990s, however, definitions of IGT and clinical tests to be performed were again discussed. This resulted in revised cutoff levels and the introduction of a new entity called impaired fasting glycemia (IFG), which is defined as high fasting glucose. Table 1 shows the new cutoff values

Table 1 Cutoff values for fasting and 2-h glucose values (mmol/l) of impaired glucose tolerance (IGT), impaired fasting glucose (IFG), and type 2 diabetes (T2D) in an oral glucose tolerance test according to guidelines by the American Diabetes Association


Whole blood




Fasting glucose 2-h glucose

>7.0 >11.1

>6.1 >10.0

>6.1 >11.1

2-h glucose




Fasting glucose




for the three diagnostic entities—IFG, IGT, and diabetes. It was also suggested that fasting glucose was sufficient for the diagnosis of glucose intolerance and type 2 diabetes.

The suggestion that a fasting sample is sufficient for the diagnosis of abnormal glycemia has been questioned, however, mainly because studies have shown that such a strategy will reduce the numbers at risk who are diagnosed and detected. This is because a large proportion of subjects with IGT have a normal fasting glucose but an elevated 2-h glucose value. In fact, there are populations with IFG alone, IGT alone, and IFG and IGT together, and these populations may represent different risks for diabetes and cardiovascular diseases. Consequently, those having a high 2-h glucose value but a normal fasting glucose, who also have increased risk for cardiovascular diseases, will be missed by the suggested strategy. A study by Larsson and collaborators from Sweden identified this dilemma since it was demonstrated that out of 414 subjects with abnormal fasting or 2-h glucose values during an OGTT, only 140 (34%) had elevation of both values. The largest group comprised subjects with high 2-h glucose values but normal fasting glucose values (i.e., IGT but not IFG), which were seen in 235 subjects (57%), whereas only 39 subjects (9%) had high fasting but normal 2-h glucose values (i.e., true IFG). The individual subgroups were shown to have similar risk factor patterns in terms of degree of obesity, blood pressure, and lipid levels. Therefore, it is now obvious that for a proper strategy to detect early cases at risk for diabetes and cardiovascular diseases, an OGTT needs to be performed since this test includes both fasting and postchallenge glucose determination.

Procedures and Evaluation of the Oral Glucose Tolerance Test

Glucose tolerance is defined as the ability to dispose a glucose load, and therefore glucose intolerance is defined as an impaired ability for glucose disposal. The gold standard technique is to challenge with an oral glucose load, with measurement of circulating glucose before and after the challenge—the OGTT. As routinely performed, this test determines the ability to dispose glucose after oral administration of 75 g glucose. The test is standardized such that it is performed in the morning after a 12-h overnight fast and blood samples are taken before the glucose load and after 2 h. Furthermore, the diet during the 3 days preceding the test should contain at least 250 g carbohydrates per day and the subjects should rest during the test in a semirecumbant position without smoking. The glucose given should be dissolved in 250-300 ml fluid, and sometimes fruit-flavored water is used to improve the taste. There has been much debate about how to take the blood sample. The original diagnostic criteria used values obtained from plasma derived from blood taken venously in tubes containing additives for prevention of coagulation. However, valid results are also obtained when glucose is measured in whole blood and when capillary samples are taken, although cutoff levels need to be adjusted for the different glucose concentrations in these samples. Arterial samples are also theoretically possible but rarely, if ever, used. Sometimes, mainly for research purposes, more frequent samples are taken and the test may last 3 h; however, for clinical purposes, the routine OGTT lasts 2 h, with a sample taken at that time point.

As shown in Figure 2, in a normal person, circulating levels of glucose increase within the first 15 min after the oral ingestion of glucose to reach a peak after 30min. Thereafter, a progressive decline occurs, with the 2-h value usually approximately 25% higher than the fasting value. Usually, it takes 3 h for a return to baseline glucose levels. In subjects with IGT, there is usually also a peak at 30 min, albeit at a higher level than in normal subjects, but the main difference versus normal subjects is that the glucose disposal is impaired, which results in a higher 2-h glucose value. In diabetics, there is usually not a peak at 30 min but a continuous rise throughout the 2-h study period. The currently used

















0 20 40 60 80 100 120

Time (minutes)

• Normal glucose tolerance (n=54) Impaired glucose tolerance (n = 16) —■— Type 2 diabetes (n = 12)

Figure 2 Venous plasma glucose levels during OGTT in subjects with normal impaired glucose tolerance, impaired glucose tolerance, and diabetes. From Ahren B, unpublished data. Means ± SEM are shown.

cutoff values are shown in Table 1. Note that the mode of measuring glucose is important with regard to the cutoff values used.

Limitations of the Oral Glucose Tolerance Test

An important limitation of the OGTT is the variability in results when the test is repeated. Actually, the coefficiency of variance (CV) is usually 15% and in some studies 20%, which is higher than that for most other clinical tests. It is not clear why OGTT has such a high CV. The variance is not, however, dependent on CV in the measurement of glucose, which is a procedure with very small error and CVs usually below 3%. Therefore, biological variation probably explains the high CV of OGTT. Factors explaining this variation may be preceding diet, exercise, emotions, stress, drugs taken for various diseases, and gender, which are all factors influencing gastric emptying, carbohydrate absorption, islet hormone secretion, hepatic glucose production, and peripheral glucose uptake (i.e., all processes contributing to the 2-h glucose value). Because of the high variability in the 2-h glucose value, a diagnosis of IGT or diabetes, particularly if intervention is planned, should not be based on a single OGTT. Instead, a clinical recommendation is to perform two OGTTs and use the mean of the two 2-h glucose values as the diagnostic value. The time interval between the two OGTTs should not exceed 3 months.

Metabolic Basis for Oral Glucose Tolerance

Oral ingestion of glucose initiates a series of metabolic perturbations, which comprise the 2-h glucose value. These metabolic perturbations are complex and involve glucose entering the bloodstream, changes in neural activity and islet hormone secretion, suppression of hepatic glucose production, and stimulation of peripheral glucose uptake. From a quantitative standpoint, of most importance with regard to the 2-h value are the changes in islet hormone secretion, which include stimulation of insulin secretion and inhibition of glucagon secretion, and the suppression of hepatic glucose production. In fact, there is an inverse linear relation between the inhibition of hepatic glucose production and the 2-h glucose value and, similarly, a linear inverse relation between stimulation of the early (first 30min) insulin secretion and 2-h glucose. This section briefly summarizes these processes.

A first series of events in the OGTT is initiated during the anticipation of the oral glucose ingestion, through olfactory stimuli and through receptors located in the oral cavity. This response is called the cephalic phase and activates sensory nerves, which give input to the central nervous system. This information is integrated in the hypothalamus for initiation and adjustment of a vagal nerve response to release insulin from the pancreatic islets. Therefore, when analyzed in detail, there is an increase in circulating insulin after glucose or meal ingestion already before glucose levels become elevated. After passage of glucose through the oral cavity, glucose passes to the stomach and through a regulated mechanism delivered into the gut. Since glucose is a monosugar, it is readily absorbed in the small intestine and reaches the splanchnic venous drainage. Glucose then passes to the portal vein and the liver. In the portal vein, glucose activates glucosensitive receptors, which through afferent sensory nerves send signals centrally to the brain for further integration with the previous signals in the hypothalamus for adjustment of efferent nerve activity. Furthermore, glucose in the liver inhibits hepatic glucose production, which is high after the overnight fast. Then, glucose passes to the general circulation to reach the pancreatic islets and the peripheral cells. The glucose load to the gut also stimulates the release of intestinal hormones, such as glucose-dependent insulinotropic polypeptide and glucagon-like peptide-1 (GLP-1). These hormones then pass through the circulation to reach the pancreas, where they stimulate insulin secretion and, as for GLP-1, inhibit glucagon secretion. In the pancreatic islets, vagal activation, intestinal hormones, and glucose stimulate insulin secretion, and glucose, GLP-1, and insulin inhibit glucagon secretion. These islet responses are of major importance for a normal glucose tolerance, and defects in these islet responses are major determinants of IGT and type 2 diabetes. Following passage of insulin into the venous drainage of the pancreas, the islet hormones reach the portal vein and the liver, and a main function of insulin is to potently suppress hepatic glucose production. This is a major process with regard to the degree of hyperglycemia during the test; in subjects with inappropriately high hepatic glucose production, the glucose level after oral glucose is high. This suppression of hepatic glucose production is augmented by the reduction in circulating levels of glucagon, which is initiated by the direct action of glucose and GLP-1 on the glucagon-producing cells and also by the action of insulin to inhibit glucagon secretion. After the liver, glucose and insulin reach the peripheral circulation and peripheral cells, where glucose is transported across the cell membranes and therefore leaves the circulation. In most cells, a major proportion of glucose uptake is sensitive to insulin; therefore, the amount of insulin, in relation to the insulin sensitivity of the cell, is of major importance for the delivery rate of glucose. However, insulin-independent mechanisms also exist, even in tissues, which are also insulin sensitive, and glucose uptake is thus also dependent on glucose. Of most importance for glucose disposal after oral glucose is the muscle cells, which have a high capacity for glucose uptake. From all these processes, the glucose level at 2 h can be determined.

It is important to realize that the metabolic processes underlying glucose tolerance are different from those underlying the fasting glucose value. Fasting glucose is mainly determined by hepatic glucose delivery during the night, which in turn is governed by the ability to maintain normal basal insulin and glucagon levels. Therefore, mechanisms underlying IFG include defective insulin secretion, defective suppression of glucagon secretion, defective sensitivity in the liver for the action of insulin, and defective peripheral glucose disposal at low glucose levels, which is a sign of insulin resistance. Although mechanisms underlying fasting and 2-h glucose values differ, there is a high correlation between fasting and 2-h glucose values in normal subjects, as shown in Figure 3. Nevertheless, there is a limited overlap between IGT and IFG in a population; in fact, most subjects with IGT have normal fasting glucose, and most subjects with IFG have a normal 2-h glucose value. This suggests that different pathophysiological processes underlie IGT and IFG, which in turn suggests that OGTT should be undertaken more frequently than performed today.

Differential Tests for Glucose Tolerance

Diagnoses of type 2 diabetes or stages preceding its occurrence can be undertaken by other means

Fasting glucose (mmol/l)

Figure 3 Correlation between fasting glucose and 2-h glucose during an OGTT in nondiabetic subjects. The regression is significant (r = 0.32, p = 0.008). From Ahren B, unpublished data.

besides OGTT. As previously stated, the use of fasting glucose has been suggested as the gold standard for diagnosis during recent years. Although it has a lower CV than the 2-h glucose value after OGTT and is simpler and more convenient for both the subject and the staff, the problem with this method is that a large number of subjects with IGT, namely those with a normal fasting glucose, will be missed.

As an alternative to OGTT, glucose tolerance may also be determined by administering glucose intravenously. In the intravenous glucose tolerance test (IVGTT), glucose is injected intravenously, usually at a dose of 0.3, 0.5, or 1 g/kg, and circulating glucose is determined before and 8, 10, 15, 20, 30, 40, 50, 60, and 80min after injection. Glucose tolerance is estimated from the elimination rate, where a glucose elimination constant (kg) is calculated. The theory behind this is that the glucose elimination after intravenous glucose displays an exponential function (i.e., after logarithmic transformation of the data, the elimination is linear. kg is thus calculated as the slope for the glucose curve following logarithmic transformation of the individual glucose values and is calculated from the formula kg = (0.693 x 100)/t1/2, where t1=2 is the half-time of glucose elimination (in minutes). The unit for kg is percentage of glucose decay per minute. Figure 4 shows this condition. Before OGTT was routinely used, IVGTT was undertaken more frequently. Unless very specific questions are asked, it is currently not used in clinical practice because it is more cumbersome to perform and it identifies only some of the metabolic processes underlying glucose tolerance, mainly insulin secretion, insulin sensitivity, and glucose uptake. Thus, other important aspects, such as glucagon secretion, release of incretin hormones, and hepatic glucose output, which are involved in the overall glucose tolerance and included in the 2-h glucose value after OGTT, contribute only marginally to the kg after IVGTT.

It has been suggested that measurement of HbA1c (i.e., the fraction of hemoglobin being glycosylated) may be used for the diagnosis of IGT and type 2 diabetes. The rationale is that hemoglobin is irreversibly glycosylated in proportion to the glucose level, and therefore the proportion of HbA1c should reflect the mean of the glucose levels during the preceding 2 or 3 months. However, although this theoretical assumption is true, measurements of HbA1c are not precise, have not been standardized at levels near the normal levels, and, consequently, slight elevations of HbA1c will not distinguish normal from impaired glucose tolerance with fairly high precision. In addition, most subjects with IGT have HbA1c levels within the normal distribution.

1 12

10 20 30 40 Time (minutes)

10 20 30 40 Time (minutes)



1.1 -








O 13













0 10 20 30 40 50 60 Time (minutes)

Figure 4 Glucose levels during an IVGTT in 41 healthy subjects with normal glucose tolerance. Glucose (0.3 g/kg) was injected intravenously at time 0. Linear regression curve for the logarithmic values from minutes 8 to 60. kg value = 1.61 ± 0.08%/min. From Ahren B, unpublished data. Means ± SEM are shown.

Clinical Aspects of IGT Epidemiology of IGT

During the 1980s, studies on the prevalence of IGT and type 2 diabetes were performed in several different populations. It became apparent that the prevalence of these conditions, although high in many populations, varied markedly between different populations. Thus, for some populations, mainly in Africa, data from only a few percent were published, whereas an exceedingly high prevalence (60%) was reported in some populations, such as Pima Indians. Figure 5 shows a collaborative study from 1993 in which data from several populations throughout the world are summarized. Studies during the past 10 years have further increased our knowledge since they have included additional populations and demonstrated that the

Mapuche Indian, Chile urban Chinese, Da Qing rural Melanesian, PNG Polish rural Polynesian, W. Samoa rural Bantu, Tanzania rural Indian, India Russian Brazilian rural Melanesian, Fiji Italian, Sanza Maltese urban Bantu, Tanzania Italian, Laurino White, USA Tunisian urban Hispanic, USA* rural Micronesian, Kiribati urban Polynesian, W. Samoa urban Indian, India urban Melanesian, Fiji rural Hispanic, USA Black, USA urban Indian, S. Africa Puerto Rican, USA urban Hispanic, USA # Chinese, Mauritius urban Hispanic, USA • rural Indian, Fiji urban Indian, Fiji urban Micronesian, Kirabati Micronesian, Nauru Pima Indian, USA

30 40

Prevalence (%)

I Diabetes mellitus

* upper income

# middle income

30 40

Prevalence (%)

I Diabetes mellitus

I I Impaired glucose tolerance

Figure 5 Prevalence (%) of diabetes and IGT in selected populations in the age range of 30-64years; genders combined. Copyright © 1993 American Diabetes Association. From Diabetes Care volume 16, page 170. Reprinted with permission from the American Diabetes Association.

prevalence of IGT and type 2 diabetes is steadily increasing. Hence, the prevalence reported in 1993 is an underestimation of the prevalence today. It has to be emphasized, however, that the difference in reported prevalence rates between different populations may be partially explained by methodological differences. For example, the prevalence of IGT and type 2 diabetes is increased by age, and in many populations there is also a higher prevalence in women than in men, at least in younger age groups. Different studies have not controlled for these confounders. Furthermore, due to migration patterns in some populations, generalization of study results is questionable, and there may also be differences in the likelihood of subjects attending a study between different populations. Nevertheless, a true ethnic difference seems to exist, with extremely high values in some Pacific island and North American Indian populations and a low prevalence in South American Indian and Bantu populations. An interesting observation is that the increase in prevalence of IGT and type 2 diabetes seems to be higher in populations with low prevalence rates and vice versa, which probably will result in diminished differences in prevalence rates between different populations in the future.

Clinical Consequences of IGT

IGT is an important risk factor for development of type 2 diabetes. However, prospective and long-term studies report different predictive values for the development of type 2 diabetes in different populations. In general, the risk of transition of IGT into type 2 diabetes ranges from 1-2% to 5% and as high as 15-20% per year. The risk is higher for those older than 50 years of age. There is also evidence that hyperglycemia, even at levels not reaching the threshold for type 2 diabetes, is associated with a substantial risk for the development of cardiovascular diseases. One explanation for this is that glucose initiates metabolic perturbations of importance for developing angiopathy, such as tissue peroxidation, production of plasminogen activation inhibition-1, and impairment of endothelial function, such as nitric oxide production. Another explanation is that hyperglycemia is associated with a number of risk factors for cardiovascular diseases, such as high blood pressure, hyperinsulinemia, dyslipidemia, and microalbuminuria, which all are included in the metabolic syndrome. In fact, if hyperglycemia is present, the risk for developing cardiovascular diseases for each of the other risk factors is augmented. Attempts to define cutoff values of glucose for cardiovascular risks have been problematic, however, probably due to the fact that the risk is continuously increased across the glucose ranges. Hence, the use of defined cutoff values is more a convenient practical issue, which is important in a clinical setting, but offers limitations from a theoretical standpoint.

Since IGT is a risk factor for type 2 diabetes and cardiovascular diseases, it is also a risk factor for overall mortality. Alberti and coworkers attempted to quantify this by performing a meta-analysis on 13 prospective studies, and they identified a hazard ratio of 1.34 (95% confidence interval, 1.14-1.57) by comparing subjects with IGT to those with normal glucose tolerance. The hazard ratio is higher for subjects with IGT than for subjects with IFG, suggesting that the 2-h glucose value is more predictive of mortality than the fasting glucose value. This shows that an individual with IGT has an increased risk not only for type 2 diabetes but also for cardiovascular diseases and hence mortality. This indicates that attempts should be made to prevent IGT from progressing to cardiovascular diseases and type 2 diabetes.

Treatment of IGT

During recent years, the issue of whether IGT may be treated to prevent progression to type 2 diabetes or cardiovascular diseases has gained considerable interest. On the one hand, it has been argued that it is important to prevent progression of IGT. On the other hand, it has been argued that treating such a large population group as those with IGT would be risky. Table 2 lists criteria that need to be fulfilled to justify prevention of a condition. In fact, most of these criteria are met for IGT; therefore, it may be argued that treating IGT is now justified. The optimal preventive intervention for IGT is not known, however. The intervention may include lifestyle changes, notably increased physical activity and dietary regulations. Such interventions have been shown to be efficient in highly motivated populations and study centers. However, whether generalization of these results to the general population is possible is not known. A clinical experience is that the outcome of advice on lifestyle changes is often disappointing in the long term. Another mode of

Table 2 Criteria for recommending population-based intervention for preventing a disease3

Criterion 1: The disease (IGT and type 2 diabetes) should pose a major health problem.

Criterion 2: Early development and natural history of the disease (IGT and type 2 diabetes) should be understood to identify parameters that measure its progression.

Criterion 3: Tests should exist for diagnosing the presumptive population (OGTT).

Criterion 4: Preventive methods should be safe, efficient, and reliable.

Criterion 5: Effort to find subjects and cost of intervention should not be burdensome and should be cost-effective.

aBased on recommendations from the American Diabetes Association (2004).

IGT, impaired glucose tolerance; OGTT, oral glucose tolerance test.

intervention is pharmacological treatment using compounds to stimulate insulin secretion, suppress hepatic glucose production, and/or enhance insulin sensitivity. These may be efficient, perhaps more efficient than advice on lifestyle changes, but may in turn pose other questions concerning long-term efficiency and potential adverse events. These two strategies are not mutually exclusive, however, and introducing pharmacological intervention without giving lifestyle advice is not appropriate in a clinical setting.

Recently, interesting data from large population studies on the prevention of progression of IGT have been obtained. Two studies, the Finnish diabetes prevention study and the Diabetes Prevention Program, have shown that lifestyle changes (i.e., individualized diet and exercise counseling) in subjects with IGT reduced the incidence of diabetes by more than 50%. In addition, in the Diabetes Prevention Program, it was shown that metformin (which reduces glucose output from the liver) reduces the risk by approximately 30%. This suggests that pharmacological treatment of IGT prevents development of type 2 diabetes. Several large studies are ongoing and results are expected within a few years.

Whether interventional programs on IGT are valid also for the prevention of cardiovascular diseases is not clearly established, mainly because long-term studies have not been performed. The STOP-NIDDM study, however, showed that acarbose, which reduces glucose absorption from the gut, reduced cardiovascular events by more than 30% during a 3-year study period. This suggests that cardiovascular diseases may be prevented by treating IGT. It should be noted, however, that for prevention of cardiovascular diseases and mortality, more studies and longer follow-up periods are required.

Quick Permanent Weight Loss

Quick Permanent Weight Loss

A Step By Step Guide To Fast Fat Loss. Do you ever feel like getting rid of the extra weight of your body? If you do, it‟s quite normal because

Get My Free Ebook

Post a comment