What are the criteria for diagnosing diabetes?

Diagnostic Criteria for Diabetes Mellitus

Diabetes is diagnosed when any one of four criteria is met: A1C ≥6.5%, fasting plasma glucose ≥126 mg/dL, 2-hour plasma glucose ≥200 mg/dL during an oral glucose tolerance test, or random plasma glucose ≥200 mg/dL with classic hyperglycemic symptoms. 1, 2

Primary Diagnostic Thresholds

Any single criterion from the following establishes the diagnosis:

• A1C ≥6.5% (48 mmol/mol): Must be performed in a laboratory using an NGSP-certified method standardized to the DCCT assay 1

• Fasting Plasma Glucose (FPG) ≥126 mg/dL (7.0 mmol/L): Fasting is defined as no caloric intake for at least 8 hours 1

• 2-hour Plasma Glucose ≥200 mg/dL (11.1 mmol/L): During a 75-gram oral glucose tolerance test (OGTT) performed per WHO protocol 1

• Random Plasma Glucose ≥200 mg/dL (11.1 mmol/L): Only valid when accompanied by classic symptoms of hyperglycemia (polyuria, polydipsia, unexplained weight loss) or hyperglycemic crisis 1, 2

Confirmation Requirements

In the absence of unequivocal hyperglycemia with acute metabolic decompensation, diagnosis requires two abnormal test results. 1, 3 This can be achieved through:

• Same test repeated on a different day: If A1C is 7.0% and repeat is 6.8%, diabetes is confirmed 1

• Two different tests both above threshold: If both A1C ≥6.5% and FPG ≥126 mg/dL from the same or different samples, diabetes is confirmed 1

• Discordant results: When two different tests yield conflicting results, repeat the test that exceeded the diagnostic threshold 1, 3

No confirmation testing is needed when a patient presents with classic hyperglycemic symptoms or hyperglycemic crisis and random plasma glucose ≥200 mg/dL 1, 4

Critical Limitations of A1C Testing

Do not use A1C for diagnosis in conditions affecting red blood cell turnover. 1, 2 Use only plasma glucose criteria in:

• Hemoglobinopathies (sickle cell disease, sickle cell trait) 1
• Pregnancy (second and third trimesters) 1, 2
• Glucose-6-phosphate dehydrogenase deficiency 1
• Recent blood loss or transfusion 1
• Hemodialysis 1
• Erythropoietin therapy 1
• Iron-deficiency anemia 1
• HIV treated with certain drugs 1

Important caveat: African Americans may have A1C levels approximately 0.4% higher than non-Hispanic whites at similar glucose levels, though the association with complications appears similar 1. Certain genetic variants can alter A1C independent of glycemia 1.

Prediabetes Categories

Prediabetes identifies individuals at high risk who require intervention:

• A1C: 5.7-6.4% (39-47 mmol/mol) 2, 3
• Impaired Fasting Glucose (IFG): FPG 100-125 mg/dL (5.6-6.9 mmol/L) 2, 3
• Impaired Glucose Tolerance (IGT): 2-hour plasma glucose 140-199 mg/dL (7.8-11.0 mmol/L) during OGTT 2, 3

Practical Algorithm for Diagnosis

Step 1: Select initial test based on clinical context:

• A1C offers greatest convenience (no fasting required) and preanalytical stability 1
• FPG has highest specificity but requires 8-hour fast 2, 5
• Random glucose only if symptomatic 1

Step 2: If initial test is positive but patient lacks acute symptoms:

• Repeat the same test on a different day, OR
• Perform a different diagnostic test 1

Step 3: If results are near diagnostic margins (e.g., A1C 6.4%, FPG 124 mg/dL):

• Discuss symptoms with patient 1
• Repeat testing in 3-6 months 1
• Consider preanalytic variability (glucose samples must be centrifuged immediately) 1

Step 4: If A1C and glucose results are markedly discordant:

• Suspect hemoglobin variants or assay interference 2, 3
• Use only glucose-based criteria 1, 2

Common Pitfalls to Avoid

Point-of-care A1C devices: While NGSP-certified POC assays exist, proficiency testing is not mandated, making them problematic for diagnosis 1. Restrict POC A1C to monitoring, not diagnosis 3.

Glucose sample handling: FPG and 2-hour plasma glucose have significant preanalytic variability if samples remain at room temperature without prompt centrifugation 1. This is the most common cause of falsely elevated results.

OGTT preparation: Patients must consume at least 150 grams of carbohydrates daily for 3 days before testing to avoid falsely elevated glucose levels 2, 3.

Test concordance: The three glucose-based tests (A1C, FPG, OGTT) do not identify identical individuals as diabetic 1, 3. A1C at the 6.5% threshold identifies approximately one-third fewer cases than FPG ≥126 mg/dL in population screening 1.

Relative Test Performance

Based on network meta-analysis comparing diagnostic accuracy against OGTT as reference 5:

• FPG ≥126 mg/dL: Highest specificity (0.98) and positive likelihood ratio (21.94), making it the best single test when positive 5
• A1C ≥6.5%: Moderate sensitivity (0.51) and high specificity (0.96) 5
• Combined HbA1c OR FPG: Highest sensitivity (0.64) but lower specificity (0.95) 5

The American Diabetes Association guidelines prioritize all four criteria equally, though FPG demonstrates superior diagnostic accuracy in comparative studies 1, 5.