Classification and Diagnosis of Diabetes

The classification of diabetes was originally limited to only two categories called juvenile-onset diabetes mellitus, now known as type 1 diabetes mellitus, and adult-onset diabetes mellitus, now known as type 2 diabetes mellitus. This has grown to a recognition of more than 50 subcategories caused by various pathogenic mechanisms or accompanying other diseases and syndromes. The diagnosis of diabetes has evolved from physician recognition of typical symptoms to detection of ambient hyperglycemia and, thence, to the definition of excessive plasma glucose levels after an overnight fast and/or following challenge with a glucose load (oral glucose tolerance test or OGTT), and more recently, by measurement of glycated hemoglobin (A1c). Screening has uncovered a much higher prevalence of diabetes in the United States and elsewhere, as well as its enormous public health impact. Modern testing has defined individuals at high risk for the development of diabetes and pregnant women whose fetuses are at increased risk for mortality and morbidity. Type 1 diabetes results from an autoimmune attack on the pancreatic islet beta cells, manifest by autoantibodies and T cells reactive with islet cell antigens prior to and after the development of hyperglycemia. When approximately 80% of beta cells have been damaged or destroyed, insulin deficiency produces hyperglycemia and risk of ketosis. Hyperglycemia, in turn, causes osmotic diuresis resulting in frequent urination, thirst, and weight loss. Type 2 diabetes is caused by a combination of insulin resistance and relative insulin insufficiency. Insulin resistance accompanies obesity, a sedentary lifestyle, and advanced age. The pathogenetic factors of type 1 and type 2 diabetes overlap in many patients, with obesity now prevalent in children and adults. Gestational diabetes is specific for pregnancy and is a harbinger of future type 2 diabetes. Diagnostic glycemic criteria for presymptomatic diabetes have been set using diabetic retinopathy as a specific complication of the disease: A1c ≥6.5%; fasting plasma glucose (FPG) ≥126 mg/dL; or plasma glucose measured 2 hours after an OGTT (2-hour PG) ≥200 mg/dL. For patients with typical symptoms, a random plasma glucose ≥200 mg/dL is diagnostic. The 2-hour PG yields the highest prevalence and A1c the lowest. A1c is the most convenient and practical test, requiring no preparation, is analytically superior, and has the lowest intraindividual variation. It is more expensive than the FPG, but the same or less than the OGTT. The 2-hour PG is the most burdensome to the patient and has the highest intraindividual variation. Standardized measurement of A1c is not available everywhere. Confirmation of an abnormal test with the same test is recommended. Studies in various populations show inconsistency among the glycemic tests. Of people meeting the A1c criterion, 27%–98% meet plasma glucose criteria. Of people meeting plasma glucose criteria, 17%–78% meet the A1c criterion. These discrepancies occur because each test measures different aspects of hyperglycemia that may vary among patients. While the risk of future diabetes is continuously associated with plasma glucose and A1c, the areas between the upper limits of normal and the diabetes cutpoints have been called “prediabetes” or “high risk for diabetes.” These have been defined categorically as A1c 6.0%–6.4% or 5.7%–6.4%; impaired fasting glucose (IFG), FPG 100–125 mg/dL; and impaired glucose tolerance (IGT), 2-hour PG 140–199 mg/dL. A1c 6.0%–6.4% increases the odds ratio (OR) for progression to diabetes (OR 12.5–16) more than the range of 5.7%–6.4% (OR 9.2). In U.S. studies, the incidence of type 2 diabetes averages approximately 6% per year in people with IGT and can reverse spontaneously. IFG is more prevalent than IGT in the United States, though IGT rises more sharply with age. IFG increases the risk of future diabetes to various degrees in different countries, with odds ratios ranging from 2.9 to 18.5. Opportunistic screening for diabetes in health care venues, especially if targeted to persons with high-risk characteristics, e.g., obesity and older age, can be cost-effective. The lower cutpoints for prediabetes should be used if the screening is also aimed at those at high risk for developing diabetes. Indiscriminate public screening for diabetes is not yet supported by sufficient long-term benefit gained from early detection of asymptomatic diabetes, nor has its cost-effectiveness been demonstrated. However, if undertaken, a capillary blood glucose ≥120 mg/dL is an efficient screening cutpoint with relatively low cost per case detected. The major public health implication of diagnosing asymptomatic diabetes is that diabetes is a major cause of cardiovascular disease, renal failure requiring dialysis and kidney transplant, and blindness or vision-threatening retinal disease necessitating surgery or intraocular injection therapy. With appropriate targeted therapy of hyperglycemia, hypertension, and dyslipidemia, these complications can be prevented or ameliorated.