1
"Is Prediabetes a Risk Factor or Is It a Disease?"

Jacob Kohlenberg1 and Adrian Vella2

1 Fellow and Instructor in Medicine, Division of Endocrinology, Diabetes, Metabolism, and Nutrition, Mayo Clinic, Rochester, MN, USA

2 Professor of Medicine, Division of Endocrinology, Diabetes, Metabolism, and Nutrition, Mayo Clinic, Rochester, MN, USA

LEARNING POINTS

• Prediabetes is a heterogeneous condition with variable risk of progression to type 2 diabetes
• In addition to diabetes risk, it is associated with an increased risk of vascular disease.
• To date, lifestyle modification is the single most important tool for altering the natural history of prediabetes and progression to type 2 diabetes.

What is prediabetes?

Prediabetes is defined as an elevated fasting plasma glucose (FPG), and/or an elevated 2-hour plasma glucose (2-h PG) during a 75-gram (g) oral glucose tolerance test (OGTT), and/or an elevated Hemoglobin A1c (HbA1c), without meeting diagnostic criteria for overt diabetes mellitus (DM) [1]. The 2020 American Diabetes Association (ADA) Guidelines define prediabetes as impaired fasting glucose (IFG) with a FPG of 100-125 mg/dL, and/or impaired glucose tolerance (IGT) with a 2-h PG during a 75-g OGTT of 140-199 mg/dL, and/or a HbA1c of 5.7-6.4% [1]. In contrast to the ADA, the 2016 World Health Organization (WHO) Guidelines define intermediate hyperglycemia as IFG between 110-125 mg/dL and/or IGT with a 2-h PG during a 75-g OGTT between 140-199 mg/dL [2]. Unlike the ADA, the WHO does not include HbA1c as a diagnostic criterion for prediabetes.

The definitions of both prediabetes and DM have evolved in recent decades. The WHO first defined the "borderline state" in 1965 as a 2-h PG during a 50 or 100 g OGTT between 110-129 mg/dL [3]. The ADA has long recognized IGT, and its definition has undergone little change since its inception. First adopted by the ADA in 1997 and WHO in 1999, the term IFG was originally defined as FPG 110-125 mg/dL [4]. However, in 2003, the ADA revised the criteria for IFG to 100-125 mg/dL based on data from multiple studies showing that the risk of DM increases markedly at a FPG concentration > 100 mg/dL [5]. In 2010, the ADA added HbA1c as a diagnostic criterion for prediabetes because the relationship between HbA1c and the risk of retinopathy was similar to corresponding FPG and 2-h PG thresholds [6].

Rationale for the diagnostic criteria for diabetes mellitus and prediabetes

Individuals with prediabetes have abnormal glucose regulation and increased risk for developing DM type 2 (DM2) and its complications [6]. The diagnostic thresholds for DM are based on [1] the bimodal distribution of FPG and 2-h PG during an OGTT and [2] the glycemic thresholds for the development of microvascular complications, specifically retinopathy (Figure 1.1). The bimodal distribution and glycemic thresholds for the development of microvascular complications have been demonstrated in many populations including the Pima, Nauruans, South Africans, Americans, Chinese, and Egyptians [7-15]. The bimodal distribution of glucose has been used to separate individuals into two groups: those with normoglycemia and those with hyperglycemia. IFG and IGT were defined as intermediates between normoglycemia and hyperglycemia. The diagnostic thresholds for FPG, 2-h PG during a 75-g OGTT, and HbA1c are relatively concordant in discriminating between the two components of a bimodal frequency distribution and their associations with microvascular complications [4]. It is important to note that defining a lower limit of an intermediate category of FPG, 2-h PG during a 75-g OGTT, and HbA1c is somewhat arbitrary because the risk of developing DM is a continuum that extends into the normoglycemic range [6].

FIG 1.1 Histogram with superimposed composite and component curves to describe the distribution of two-hour plasma glucose levels following an oral glucose load. Glucose concentrations and frequencies were arbitrarily chosen to illustrate a bimodal distribution. The bimodal glucose distribution can be used to separate individuals into two groups, those with normoglycemia and those with hyperglycemia. Intermediate glucose concentrations between normoglycemia and hyperglycemia helped define the diagnostic thresholds for prediabetes.

Epidemiology of prediabetes

The 2020 National Diabetes Statistics Report published by the Centers for Disease Control and Prevention estimated the prevalence of prediabetes (defined by 2020 ADA criteria) to be 38.0% (95% confidence interval (CI) 35.2-40.8) among adults in the United States (U.S.) [1, 16]. Overall, the prevalence of prediabetes has not changed significantly from 2005-2016. However, the number of U.S. adults who are aware that they have prediabetes increased from 6.5% (95% CI 5.3-7.9) in 2005-2008 to 13.3% (95% CI 11.0-16.0) in 2013-2016. Further, the prevalence of prediabetes increases with age: 29.1% (95% CI 25.2-33.3) of adults 18-44 years of age; 46.3% (95% CI 43.5-49.1) of adults 45-64 years of age; and 51.0% (95% CI 46.5-55.5) of adults = 65 years of age. Prediabetes is also more common in men, whose prevalence is 42.3% (95% CI 38.1-46.5), compared to women, whose prevalence is 33.7% (95% CI 30.7-36.8). The prevalence of prediabetes is similar among racial/ethnic groups and among individuals of different education levels [16].

Pathogenesis of impaired fasting glucose and impaired glucose tolerance

In epidemiologic studies, isolated IGT consistently has a higher prevalence than isolated IFG [17]. The prevalence of IFG and IGT increases with age [17]. In adults less than 55 years of age, IGT is more common in women and IFG is more common in men [17]. This suggests that these two states have different pathophysiologic mechanisms.

Genetics and lifestyle influence the pathogenesis of DM [18, 19]. Although more than 400 genetic signals have been identified as influencing risk for DM2, single polymorphisms add only small degrees of risk [20]. Polymorphisms in the Transcription factor 7-like 2 (TCF7L2) locus have the largest-known effect on risk for DM [20, 21]. Compared with non-carriers, heterozygous and homozygous carriers of the at-risk TCF7L2 variants have relative risks of developing DM of 1.45 and 2.41, respectively [20, 21].

The relationship between genetics and lifestyle on the pathogenesis of DM was emphasized by a study comparing the prevalence of DM2 in the Pima population in Mexico versus the Pima population in the U.S. [22]. The prevalence of DM2 in the Pima population in Mexico was 6.9%, compared to 38% in the Pima population in the U.S. The prevalence of obesity in the Pima population in Mexico was significantly lower than that in the Pima population in the U.S. By comparison, the latter group also had significantly lower physical activity levels.

Overall, islet cell function is the primary regulator of glucose metabolism, but multiple additional factors contribute. Postprandial and fasting glucose concentrations are determined by insulin secretion, hepatic extraction, insulin action, glucagon suppression, glucose effectiveness, and the rate of gastric emptying (Figure 1.2) [23-29].

Studies using the minimal model to quantitate insulin secretion and insulin action demonstrated that both indices are lower in subjects with NFG/IGT and IFG/IGT than in subjects with NFG/NGT [25, 26]. However, there was no significant difference in insulin secretion and insulin action between subjects with IFG/NGT and those with NFG/NGT. This implies that insulin secretion declines in concert with insulin action across the spectrum of prediabetes; however, subjects with isolated IFG may have an altered glucose threshold for insulin secretion without other intrinsic defects in ß-cell function.

Using human pancreatic tissue obtained at autopsy, obese individuals with IFG have a 40% deficit in relative beta-cell volume compared to obese individuals with NFG [30]. Additionally, the diabetes-associated genetic variation in TCF7L2 is associated with impaired insulin secretion [31-33]. The inability of insulin secretion to compensate for a decline in insulin action results in hyperglycemia.

It is also essential to consider the role of hepatic extraction in the pathophysiology of prediabetes and DM [26]. The peripheral insulin concentration reflects portal insulin secretion, hepatic extraction, distribution, and degradation [34]. Hepatic extraction of secreted insulin has a direct relationship with disposition index. Reduced hepatic extraction in prediabetes and DM may be a compensatory measure to increase circulating insulin. Interestingly, insulin pulse characteristics influence hepatic extraction of insulin, and the diabetes-associated genetic variation in TCF7L2 is associated with abnormal insulin pulse characteristics [33].

Furthermore, lack of glucagon suppression contributes to hyperglycemia in subjects with impaired insulin secretion [24]. The...