Clinical Dilemmas in Diabetes

Adrian Vella, Robert A. Rizza, Adrian Vella, Robert A. Rizza

• Publisher: Wiley-Blackwell
• Year: 2011
• ISBN:
  9781444340273
• Edition: 1
• Topic: Medizin
• Subtopic: Endokrinologie & Stoffwechsel

Clinical Dilemmas in Diabetes provides evidence-based clinical guidance on the most common and problematic areas of concern encountered in diagnosing, treating and managing patients with diabetes. Each chapter is highly topical and has been selected due to current interest, specific recent developments, and areas of controversy.

This valuable guide provides assistance in managing the life-long treatment of diabetes and the complications that often develop in patients. Clinical Dilemmas in Diabetes guides the medical team in their decision-making, particularly when there are conflicts in the treatment for the disease and the complications.

Part of the Clinical Dilemmas series, the well-focused chapter structure allows for quick retrieval of information, and each opens with a "Learning Points" box to aid easy assimilation of the main issues. With a leading team of contributors and editors, Professor Robert A. Rizza is the immediate Past-President of the American Diabetes Association.

This book is perfect for use on the wards and clinics as well as for self-study by diabetologists, diabetes specialist nurses, endocrinologists, GPs and cardiologists.

PART I

Prediabetes and the Diagnosis of Diabetes

1

Is prediabetes a risk factor or is it a disease?

Kalpana Muthusamy1 and Adrian Vella2

1Clinical Fellow, Division of Endocrinology, Mayo Clinic Rochester, MN, USA

2Associate Professor of Medicine, Department of Endocrinology, Mayo Clinic, Rochester, MN, USA

LEARNING POINTS

• The diagnostic criteria for prediabetes and diabetes are based on the relationship of hyperglycemia with microvascular disease.
• Defects in insulin secretion and action occur in people with impaired fasting glucose and impaired glucose tolerance.
• An oral glucose tolerance test may help to better characterize patients at higher risk of progression to type 2 diabetes.
• Intervention may delay the progression to diabetes.

Prediabetes, as previously defined by the American Diabetes Association (ADA), includes subjects with fasting plasma glucose (FPG) >100 mg/dl and <126 mg/dl and/or 2-hour plasma glucose following a 75-g oral glucose load >140 mg/dl and <200 mg/dl. The rate of progression to diabetes without any intervention is about 28.9% over a 3-year period as seen in the placebo arm of the Diabetes Prevention Program [1]. A 9-year longitudinal study from Olmsted County, Minnesota, reported a similar rate of diabetes progression of 34% [2]. The prevalence rate of prediabetes in the American adult population, as reported by CDC from 2003 to 2006, was 25.9% [3]. This represents 57 million American adults, a significant number of whom are predisposed to developing diabetes if adequate intervention is not undertaken. Therefore, understanding the definition of prediabetes, its implications, pathogenesis, and appropriate management becomes critical to any clinician.

Prediabetes includes two categories, impaired fasting glucose (IFG) and impaired glucose tolerance (IGT). Examining the evolution of these criteria will help us understand not only the basis of the current definitions, but also provide us guidance for the necessary evaluation and management.

Prediabetes, diabetes, micro- and macrovascular disease

Impaired glucose tolerance (IGT) is defined by a plasma glucose 2 hours after a 75-g oral glucose load >140 mg/dl and <200 mg/dl, while impaired fasting glucose (IFG) is defined by a fasting plasma glucose >100 mg/dl and <126 mg/dl.

IGT is a terminology that has been long known and has been a part of the ADA classification since 1979 [4]. IFG as a separate entity was established in an ADA report published in 1997 [5] and was later adopted by an expert WHO panel in 1999 [6]. These categories were intended to be seen as risk factors for future diabetes and cardiovascular disease rather than distinct clinical groups. The definition for IGT has undergone little change since its inception. IFG was initially defined as fasting plasma glucose >110 mg/dl and <126 mg/dl. This classification was rather arbitrary and reflected the then available evidence suggesting an insulin secretory defect and an increased risk of cardiovascular disease.

Brunzell et al. performed intravenous glucose tolerance tests in 66 subjects with a wide range of fasting glycemia [7]. Acute insulin response and glucose disappearance rate were markedly lower in subjects with fasting plasma glucose above 115 mg/dl in comparison to those with a fasting glucose below 115mg/dl. The main limitation of this data was the relatively small number of patients in the fasting plasma glucose group 115–149 mg/dl (n = 3). The Paris Prospective Study noted an increasing risk of diabetes with incremental fasting plasma glucose, despite normal glucose tolerance (2-hour-value after a 75-g oral glucose tolerance test <140 mg/dl). The relative risk of developing subsequent diabetes in the IGT and IFG (fasting plasma glucose >109 mg/dl) groups was 9.6 and 5.6, respectively [8]. The impact of hyperglycemia on cardiovascular mortality was also examined in this study, with the age-adjusted relative risk for coronary heart disease death noted to be 1.32 (1.04–1.67) in subjects in the fasting plasma glucose category of 104–124 mg/dl in comparison to the group <104 mg/dl [9]. A similar increase in risk was observed with increased post-challenge glucose in the Whitehall study that followed 18,403 male civil servants for a total of 7.5 years [10].

The microvascular effects of prediabetes were investigated in a few studies with mostly uniform results. Subjects with a capillary blood glucose between 120 and 200 mg/dl, following a 50-g oral glucose load did not have any discernible difference from controls in the prevalence of retinal abnormalities over a 10-year follow-up period [11]. Klein et al. evaluated the effect of impaired glucose tolerance, with plasma glucose between 140 and 200 mg/dl after a standard 75-g oral glucose load [12]. Age-adjusted frequency of visual impairment as measured by visual acuity of ≤ 20/40 was higher in the IGT group when compared to men with diabetes and normoglycemic women. However, the rates of retinopathy were uniformly low across all groups with no significant intergroup differences. In another report from two different groups of patients, including Pima Indians and male civil servants, development of retinopathy was mostly confined to subjects with 2-hour plasma glucose exceeding 200 mg/dl, without any marked change in the intermediate groups [13].

Following these earlier studies, one of the important debates that ensued was the comparability between IGT and IFG with regard to outcomes. Data from a longitudinal study of Pima Indians showed greater prevalence of IGT over IFG among nondiabetic subjects [14]. However, the 5-year cumulative incidence of diabetes was much higher for IFG at 31% in comparison to 19.9% for subjects with IGT. The combination of these two risk factors was better than either alone with an incidence of 41.2%. A receiver operating characteristic (ROC) curve analysis showed that, by defining IFG using a fasting glucose ≥ 102 mg/dl, the prevalence in the two groups was mirrored. This might not have necessarily led to identifying the same set of subjects, as these two cohorts might have included subjects who were mutually exclusive. However, the sensitivity and specificity of diabetes prediction was equaled in the IFG and IGT groups when using a definition of IFG >103 mg/dl as opposed to 110 mg/dl. In a Mauritian [15] cohort of 3,229 nondiabetic subjects, 148 had IFG alone in comparison to 489 with isolated impaired post-challenge glucose. A combination of IFG and IGT was present in 118 subjects. The sensitivity, specificity, and positive predictive value for prediction of progression to diabetes were 50, 84, and 24%, respectively, for IGT. Although IFG was less sensitive, it had a better specificity and positive predictive value at 26, 94, and 29%, respectively. These data would suggest that IFG defines a smaller, yet a more extreme category of glycemia that progresses to diabetes more predictably. However, from a population perspective, IFG identifies a lesser percentage of people progressing to diabetes, making it difficult to successfully implement diabetes prevention measures based on fasting plasma glucose alone. In this study, the optimal definition of IFG that gave the best combination of sensitivity and specificity for diabetes prediction was a fasting glucose >99 mg/dl [16]. These data formed the basis for the revised IFG criteria of plasma glucose >100 mg/dl and less than 125 mg/dl, in a follow-up report in 2003 [17].

It is important to remember that in clinical practice, the risk of progression to diabetes follows a gradient across a seamless continuum of glucose levels [16]. While scrutinizing the evidence to decide the optimal definitions of IGT and IFG and their individual value, it is critical to understand if IGT and IFG act as risk factors for micro- and macrovascular disease independently of diabetes.

It is generally accepted that microvascular disease, such as retinopathy, neuropathy, and nephropathy, is a function of the degree and duration of hyperglycemia. Contrary to the earlier studies that did not reveal an increased frequency of retinopathy among people with prediabetes, some recent studies demonstrated an elevated risk of microvascular disease even in subjects with hyperglycemia less than the diabetic range. A subset of the Diabetes Prevention Program cohort was investigated with fundus photographs at a mean 5.6 years of follow-up [18]. Changes of diabetic retinopathy were reported in 7.9% of the impaired glucose group and in 12.6% of the group that developed diabetes on follow-up. Although the subjects who developed retinal changes were not significantly different from those without these changes in the impaired glucose group, they tend to have a higher baseline prevalence of hypertension, lower HDL, higher triglycerides, and a history of gestational diabetes. The rates of retinopathy and nephropathy were higher in individuals with impaired fasting glucose in comparison to those with impaired glucose tolerance on 10 years of follow-up of a group of Pima Indians, also supporting the previous evidence that IFG might denote a metabolically advanced state [19]. As opposed to these results, the incidence of diabetic retinopathy was reported to be very low at 28–31/10,000 person-years of follow-up in a large Japanese cohort of atomic bomb survivors with impaired glycemia [20]. A steep rise in the incidence and prevalence of fundus changes were noted only when the fasting plasma glucose was >125 mg/dl and the 2–hour post-challenge glucose >198 mg/dl. A similar threshold for retinopathy also evolved in the AusDiab study [21]. A clear threshold effect was not evident for microalbuminuria and the relation to rise in glucose was more gradual. Subjects with neuropathy were more likely to have retinopathy and microalbuminuria in the AusDiab cohort with impaired glucose metabolism [22]. Collectively, although there is evidence for increased prevalence and incidence of microvascular changes before the onset of diabetes, these changes predominantly occur with higher levels of glycemia.

In summary, the recently proposed definitions of prediabetes are dependent on their ability to identify individuals with a high risk of progression to diabetes. Defining IFG using a fasting glucose >100mg/dl increased the prevalence of prediabetes from 19.3% to 36.3% on evaluation of the NHANES III data [23]. Whether this definition portends true benefit or places a higher societal burden for preventive measures has been questioned [24]. We also have to factor in the behavioral impact of this labeling on individuals [25]. Strong antagonistic opinions to the new cutoff cite the lack of net proven benefit based on a detailed decision analysis [26].

Most recently in 2010, the title “Prediabetes” was renamed as “Categories of increased risk for diabetes” to reflect the risk of progression to diabetes rather than the subsequent micro- and macrovascular outcomes. An equivalent intermediate category for A1C was also identified, with values between 5.7% and 6.4% indicating a heightened risk for diabetes development [27–29].

Prediabetes and atherosclerosis: Why do they associate and how to best predict the risk?

A progressive increase in cardiovascular risk has been shown with rising blood sugars, across a spectrum ranging from normal to significant hyperglycemia. The DECODE study group showed a J-shaped relationship between all-cause mortality and plasma glucose, whether fasting or post-challenge [30]. A plausible and intuitive explanation for the increased cardiovascular risk is the clustering of other well-known traditional risk factors in patients who develop prediabetes [31]. The San Antonio Heart Study followed 614 nondiabetic Mexican American individuals and demonstrated that subjects who developed diabetes had a more atherogenic profile at baseline, including higher triglycerides, LDL and total cholesterol, BMI, blood pressure, insulin, and lower HDL than the group that did not develop diabetes [32]. The clustering of risk variables explained all the observed metabolic features rather than a single underlying etiology [33]. Low cardiorespiratory fitness had a significant impact on all-cause mortality in women with IFG in the 16-year follow-up in the Aerobic Center Longitudinal Study (ACLS) [34].

Studies that examined the ability of IGT and IFG to predict cardiovascular risk and mortality suggest that IGT is a better predictor of all-cause mortality [35,36] and cardiovascular disease [37–39]. In contrast, data from Norwegians followed over 22 years showed that fasting plasma glucose was an important predictor of cardiovascular death [40]. Adding to these already varied results, a Chinese study showed equivalent performance of IGT and IFG in predicting cardiovascular disease risk [41]. The Atherosclerosis Risk in Communities Study (ARIC) also showed that both IGT and IFG were associated with an increased prevalence of cardiovascular risk factors with none being worse than the other [42]. It is also important to remember the important role of other cardiovascular risk factors in the development of atherogenesis. In agreement, the Framingham Offspring and San Antonio Heart Studies have shown that the knowledge we gain from post-challenge hyperglycemia might add little to what we might already know from traditional cardiovascular risk factors [43,44].

Is there a role for OGTT in clinical practice?

T...