Diagnostic criteria for type 1 diabetes

The diagnosis of type 1 diabetes is made based on the same four biochemical criteria valid for all forms of diabetes. The first criterion is glycated hemoglobin (HbA1c) ≥6.5% (48 mmol/mol). The second is fasting plasma glucose ≥126 mg/dL (7.0 mmol/L), after at least 8 hours without caloric intake. The third is 2-hour plasma glucose during the oral glucose tolerance test (OGTT) with 75 g of glucose ≥200 mg/dL (11.1 mmol/L). The fourth is a random plasma glucose ≥200 mg/dL (11.1 mmol/L) accompanied by classic symptoms of hyperglycemia or by a hyperglycemic crisis, such as ketoacidosis [1].

Type 1 diabetes frequently appears at young ages, often begins suddenly, with intense thirst, frequent urination (including at night), weight loss and sometimes ketoacidosis. To these biochemical criteria can sometimes be added (if sought) evidence of autoimmunity (specific autoantibodies) and of endogenous insulin deficiency (low C-peptide). These additional investigations do not change the logic of using the diagnostic criteria. They sometimes help the doctor distinguish between type 1 and type 2 or other forms of diabetes. They are useful above all when the clinical picture is not clear, as the figure below shows [1].

The general rule requires two criteria with abnormal results for a firm diagnosis of diabetes. These can come either from the same test (criterion) repeated on a different date, or from two different tests collected on the same day. A common pair is HbA1c and fasting plasma glucose. If both results are above the threshold, the diagnosis of diabetes is confirmed. If one is positive and the other is not, the test with the abnormal value is repeated or another test is used. This rule reduces the risk of laboratory errors and of misdiagnosis, though it does not remove that risk entirely [1].

There is one clear exception. If you have classic symptoms of hyperglycemia or present with a hyperglycemic crisis, and the random plasma glucose is ≥200 mg/dL (11.1 mmol/L), a single test is sufficient for diagnosis. This exception applies very often in type 1 diabetes. If you arrive at hospital with ketoacidosis, very high blood glucose values and obvious symptoms, the doctor does not wait for a second test, but immediately starts insulin treatment. In these cases, delaying treatment initiation would be dangerous, and the diagnosis is clear (with the exception of neonatal diabetes, a rare, genetically caused form that appears in the first months of life) [1].

Glycated hemoglobin reflects your average blood glucose over the last 2–3 months. Glucose in the blood binds slowly and irreversibly to hemoglobin in red blood cells, and these red blood cells live approximately 120 days. An HbA1c value ≥6.5% (48 mmol/mol) shows that blood glucose has been consistently elevated for a long enough period to establish the diagnosis of diabetes. This threshold was chosen because it marks the level from which the risk of chronic complications increases significantly, especially diabetic retinopathy [2].

In the context of type 1 diabetes, however, HbA1c has some limitations. If onset was rapid, within a few weeks, glycated hemoglobin may still be normal or only slightly elevated, even though your current blood glucose is far above the limit. That is why, in acute presentations, the doctor does not wait for HbA1c, but relies on blood glucose and symptoms. Furthermore, HbA1c cannot be used for diagnosis in certain situations such as anemia, hemoglobinopathies (thalassaemia or sickle cell disease), pregnancy, hemolysis, recent transfusions or kidney failure [1] [2].

The oral glucose tolerance test (OGTT) measures how your body processes a standard sugar load. You present in the morning, after a fast of at least 8 hours, and your baseline blood glucose is collected. Then you consume 75 g of glucose dissolved in water, and at 2 hours the second sample is collected. The interpretation of the 2-hour value is as follows:

• below 140 mg/dL (7.8 mmol/L) — normal glucose tolerance;
• between 140 and 199 mg/dL (7.8–11.0 mmol/L) — impaired glucose tolerance, that is, prediabetes;
• ≥200 mg/dL (11.1 mmol/L) — confirms diabetes [1].

In type 1 diabetes, the OGTT is rarely used in routine practice. Usually, when you reach the doctor, blood glucose is already very high and the symptoms are obvious. The diagnosis is made directly, without further need for OGTT testing. The OGTT becomes useful in atypical situations, for example in slowly progressive forms of autoimmune diabetes in adults. It is also used in monitoring programs for first-degree relatives with positive autoantibodies, in the presymptomatic stages (1 and 2) of type 1 diabetes [3]. In these situations, the OGTT helps to track disease progression before symptoms appear.

A random plasma glucose ≥200 mg/dL (11.1 mmol/L) is sufficient for the diagnosis of diabetes only under certain conditions. It must be accompanied by the classic symptoms of hyperglycemia or by a hyperglycemic crisis, such as ketoacidosis. Classic symptoms include polyuria (frequent and abundant urination), polydipsia (intense thirst), polyphagia (excessive hunger) and unexplained weight loss. In this situation, the diagnosis of diabetes can be made on the basis of a single measurement, without the need for confirmation by a second test [1].

If a random plasma glucose is ≥200 mg/dL (11.1 mmol/L) but you have no symptoms, the diagnosis is not made on this result alone. In this case the doctor evaluates the other diagnostic criteria (fasting plasma glucose, HbA1c, OGTT). In type 1 diabetes, however, presentation is rarely asymptomatic. Most often the classic symptoms have been present for several weeks, and patients often present to the doctor directly with diabetic ketoacidosis [4].

Autoantibodies are markers that show your immune system is attacking the beta cells in the pancreas, the ones that produce insulin. The four standard autoantibodies tested for type 1 diabetes are antibodies against glutamic acid decarboxylase (GAD65), insulin autoantibodies (IAA), tyrosine phosphatase antibodies (IA-2), and zinc transporter 8 antibodies (ZnT8). There are also islet cell antibodies (ICA), but they are not reliable enough to be used for this purpose at present. The presence of at least two autoantibodies confirms the autoimmune origin of the disease and defines autoimmune type 1 diabetes, called type 1A. The form without autoantibodies is called type 1B or idiopathic [5].

Autoantibody testing may be recommended at the time of diagnosis, especially when the presentation is atypical or when the doctor has doubts between type 1 and type 2. In adults, anti-GAD65 antibodies are the most frequently positive and may persist for many years after onset. In children and adolescents, IAA and IA-2 are often the first to appear. As the years pass since onset, some autoantibodies may decrease or disappear, but this does not change the diagnosis. The presence of autoantibodies classifies diabetes as type 1, regardless of the age at onset, including in slowly progressive adult forms (LADA) [6].

C-peptide is a molecule released by the pancreas in amounts equal to your own (endogenous) insulin. When you measure C-peptide, you are actually measuring how much of your own insulin you still produce. In type 1 diabetes the values are low or undetectable because the beta cells are destroyed by the autoimmune process. In type 2 diabetes C-peptide is usually normal or sometimes even elevated because the pancreas is still producing insulin, but the tissues are resistant to its action. For correct interpretation, C-peptide must be measured at the same time as blood glucose. That tells the doctor how strongly the endocrine pancreas was stimulated to secrete insulin [7].

C-peptide is not used to make the diagnosis of diabetes, but rather to guide towards its type. It is useful in atypical cases, for example in an adult with normal weight and rapid onset. In such patients the doctor hesitates between type 1 diabetes, type 2 diabetes and rare forms such as monogenic diabetes. It can also sometimes help confirm the slowly progressive autoimmune form in adults (LADA), where values are intermediate and decrease gradually over a few years [6]. During the remission period, also called the honeymoon, C-peptide may even be close to normal for a while (in total remission) [7]. That means your pancreas is producing a significant amount of insulin.

No. Neither ketonaemia (ketone bodies in the blood) nor ketonuria (ketone bodies in the urine) are part of the official diagnostic criteria for diabetes. The diagnosis of type 1 diabetes is made exclusively on the basis of blood glucose or glycated hemoglobin. Ketone bodies appear when the body can no longer use glucose due to lack of insulin and begins to break down fats for energy. The result is the ketone bodies (acetone, acetoacetate and beta-hydroxybutyrate), which accumulate in the blood and are eliminated through urine [1].

Although they are not diagnostic criteria, the presence of ketone bodies is an extremely important sign for you. Ketone bodies indicate a severe insulin deficiency and argue for the presence of type 1 diabetes rather than type 2 diabetes, especially at onset. In diabetic ketoacidosis, the combination of hyperglycemia, increased ketone bodies and metabolic acidosis constitutes a medical emergency that requires immediate presentation to the emergency room [4]. Therefore, ketone bodies do not make the diagnosis, but they orient you towards type 1 diabetes and contribute to estimating the severity of the situation. If you already have a diagnosis of type 1 diabetes, monitoring ketone bodies on sick days or with very high blood glucose values is a very important rule.

The numerical thresholds are identical in children and adults. The same four diagnostic criteria apply at any age. The only particularity is that in children the glucose tolerance test uses 1.75 g of glucose per kilogram of body weight. The dose never exceeds 75 g [1].

What differs is the clinical picture and the frequency of certain markers. In children and adolescents, the onset of type 1 diabetes is usually rapid, and the classic symptoms appear within a few weeks. Ketoacidosis at presentation is frequent, and autoantibodies are usually present in greater numbers [8]. In adults, the presentation can be more misleading and may sometimes resemble type 2 diabetes, especially if you are overweight. The slowly progressive autoimmune form in adults (LADA) is easily confused with type 2 diabetes in the first years, until insulin reserve decreases significantly. Therefore, although the criteria are the same, clinical context, autoantibodies and C-peptide are more important in adults to establish the correct type of diabetes [9].