Monday , November 20 2017
Home / Conditions / Type 1 Diabetes / International Textbook of Diabetes Mellitus, 4th Ed., Excerpt #6: Classification of Diabetes Mellitus and Other Categories of Glucose Intolerance Part 6 of 6

International Textbook of Diabetes Mellitus, 4th Ed., Excerpt #6: Classification of Diabetes Mellitus and Other Categories of Glucose Intolerance Part 6 of 6


Impaired glucose regulation (impaired glucose tolerance and impaired fasting glycemia)

Impaired glucose tolerance (IGT) and impaired fasting glycemia (IFG) are categorized as stages in the natural history of disordered carbohydrate metabolism. They occur in all individuals as they progress from normal to diabetes, but since the transition through these states is rapid in type 1 diabetes, they are rarely identified in such individuals. Therefore, nearly all of the literature dealing with IGT and IFG is concerned with issues relating to type 2 diabetes, such as risk of developing type 2 diabetes and CVD.

IFG and IGT represent ametabolic state intermediate between normal glucose homeostasis and diabetes. The pathophysiologic aspects of hyperglycemia of each category are somewhat different. IGT is associated with muscle and liver insulin resistance and thus IGT is often associated with the metabolic or insulin resistance syndrome [7], while IFG is usually related to insulin secretory deficits.

A meta-analysis suggested that there is a positive relationship between IFG/IGT and diabetes which varies across ethnicity and age [83]. This study showed that individuals with combined IGT and IFG had the highest risk of future diabetes. In terms of sensitivity and specificity for the subsequent development of diabetes, the sensitivity of IFG as originally defined at 6.1 mmol L−1 (110mg dL−1) is less than that of IGT in most populations [84], but the specificity of IFG is greater [85]. IGT is more common than IFG using 6.1mmol L−1 (110mg dL−1) in most populations but it should be noted that the sensitivity and specificity of both IGT and IFG is entirely dependent on the cutpoints selected, and not on any inherent differences between FPG and 2hPG [86]. If IGT and IFG are defined such that they have similar prevalence to each other, they then have the same predictive values for subsequent diabetes [86]. Further, with the ADA cutpoint of IFG (5.6mmol L−1 or 100mg dL−1—see later), the sensitivity of IFG is similar to IGT, but the specificity falls.

Thus, neither the risk of developing diabetes nor the sensitivity and specificity for future diabetes seem to differ enough between IGT and IFG to suggest one category is more useful than the other. In reality, in most populations IGT is more prevalent than IFG (if IFG is defined as FPG of 6.1–6.9 mmol L−1 (110–125mg dL−1)), and thus it identifies a greater proportion of those who will develop diabetes. Furthermore, although at the lower cutpoint of IFG of FPG 5.6–6.9 mmol L−1 (100–125mg dL−1), the prevalence of IFG approaches that of IGT, the two groups remain limited in their overlap. Relying on only a FPG will not identify the same proportion of individuals at risk compared to undertaking an OGTT.

The relationship between IGT and IFG and CVD is well studied in meta-analyses. In a review of the evidence from 27 studies [87], IFG (at both cutpoints) and IGT were both associated with a significantly increased risk of approximately 20% of CVD.

Diagnosis of IGT and IFG categories has been traditionally made by measuring blood glucose levels, either in the fasting state (for IFG) or during an OGTT (for IGT) (see Table 1.6 for cutpoints). Since individuals with IFG may have diabetes, it is recommended that those who are found to have IFG should have an OGTT to exclude diabetes [7].

Whilst IGT has been part of the classification of glucose intolerance for many years, IFG was only added in 1997, with a lower cutpoint of 6.1mmol L−1 (110mg dL−1). However, in 2002, the ADA proposed a new cutpoint of IFG of 5.6mmol L−1 (100mg dL−1), as this maximized the sensitivity and specificity for predicting future diabetes [88]. On review of the same evidence, the WHO decided not to adopt this new cutpoint, as it significantly increased the number of people being labeled as abnormal, but without evidence that so doing would improve outcomes [69].

The purpose of defining other categories of glucose intolerance or prediabetes is to identify a group of the population at increased risk for the development of both diabetes and CVD, so that interventions (lifestyle and pharmacologic) can be applied to reduce these risks. IGT and IFG are considered risk factors for diabetes and CVD.

In summary, longitudinal data show that IFG and IGT are rather similar to each other in their ability to predict future diabetes and CVD. However, since the populations of IFG and IGT have limited overlap with each other, undertaking the OGTT to identify those with IGT provides the opportunity to identify a greater proportion of the at-risk population.


The notion underpinning setting a normal category of glucose is that people with values below the upper limit of normal are at no or only “normal” risk of developing diabetes or its micro- and macrovascular complications [5,7,84,89]. Since the risks of future development of diabetes and CVD are related to blood glucose across most of its spectrum, and well into any normal ranges that have been set, such notions of “normal” blood glucose should be interpreted very cautiously. The actual setting of the cutpoint indicating normoglycemia over the years has undergone considerable changes. The early classification in 1985 by US NDDG that was adopted by WHO had set diabetes at a fasting glucose of 7.8mmolL−1 (140mg dL−1) and those under this threshold were labeled “normoglycemia”. In 1997, upon the availability of new data, the ADA, with support from WHO reset the cutpoint of a “normal” fasting plasma glucose from 7.8mmol L−1 (140mg dL−1) to 6.0mmol L−1 (110mg dL−1) [5,7,89]. In 2002–2003, the ADA recommended that the cutpoint be 5.5mmol L−1 (100mg dL−1) [88].


The classification of diabetes is an evolving process. As the research into diabetes is a continuing and dynamic process and epidemiologic and clinical studies are in progress, there may well be revision and refinement of the classification system. This is especially important given the recent recommendation to use HbA1c to diagnose diabetes and the caveats to its use. As more knowledge emerges about the etiology of cases currently positioned in the type 2 process category, modification and refinement may be necessary.

Click here for references: