Distinguishing Type 1 from Type 2 Diabetes: Laboratory Testing
The Tests You Need
To distinguish type 1 from type 2 diabetes, measure islet autoantibodies first (GAD65, IA-2, ZnT8, insulin antibodies), and if antibody-negative or results are ambiguous, proceed to C-peptide testing—preferably a random sample within 5 hours of eating. 1, 2, 3
Autoantibody Testing: First-Line Approach
Test for glutamic acid decarboxylase (GAD65) antibodies, islet tyrosine phosphatase 2 (IA-2) antibodies, zinc transporter 8 (ZnT8) antibodies, and insulin antibodies to confirm autoimmune etiology. 1, 2, 4
If any autoantibody is positive, the diagnosis is definitively type 1 diabetes regardless of C-peptide level, though C-peptide should still be low. 2, 5
GAD65 is the most frequent marker in both type 1 and type 2 phenotypes and is highly predictive for eventual insulin requirement. 4
Approximately 5-10% of adults with type 1 diabetes are antibody-negative, making C-peptide measurement essential in this subset. 2, 5
C-Peptide Testing: When and How
Timing and Methodology
A random C-peptide sample within 5 hours of eating can replace formal stimulation testing for diabetes classification. 2, 3, 5
For fasting C-peptide measurement, ensure simultaneous fasting plasma glucose is ≤220 mg/dL (12.5 mmol/L). 2, 3
Never test C-peptide within 2 weeks of a hyperglycemic emergency (DKA) as results will be artificially suppressed. 2, 3, 5
If concurrent glucose is <70 mg/dL (<4 mmol/L), repeat the test as hypoglycemia artificially lowers C-peptide. 2, 3
Interpretation of C-Peptide Results
Type 1 Diabetes:
C-peptide <200 pmol/L (<0.6 ng/mL) is consistent with type 1 diabetes and indicates significant beta-cell loss. 2, 3, 5
C-peptide <80 pmol/L (<0.24 ng/mL) indicates absolute insulin deficiency and severe beta-cell loss—this result does not need repeat testing. 2, 5
Indeterminate Zone:
C-peptide 200-600 pmol/L (0.6-1.8 ng/mL) may indicate type 1 diabetes, LADA (latent autoimmune diabetes in adults), MODY (maturity-onset diabetes of the young), or long-standing insulin-treated type 2 diabetes. 2, 3, 5
In this range, check for MODY features: age <35 years, HbA1c <7.5% at diagnosis, one parent with diabetes, and absence of autoantibodies. 5, 6
Type 2 Diabetes:
C-peptide >600 pmol/L (>1.8 ng/mL) strongly suggests type 2 diabetes regardless of testing circumstances. 3, 5
Look for supporting features: BMI ≥25 kg/m², absence of weight loss, absence of ketoacidosis, and features of metabolic syndrome. 5
Clinical Algorithm for Ambiguous Cases
Step 1: Measure islet autoantibodies (GAD65, IA-2, ZnT8) in overweight/obese adolescents or adults with unclear diabetes type. 1
Step 2: If antibody-positive → Type 1 diabetes confirmed; initiate basal-bolus insulin or insulin pump therapy. 1, 2
Step 3: If antibody-negative → Measure C-peptide (random within 5 hours of eating or fasting with glucose ≤220 mg/dL). 2, 3
Step 4: Interpret C-peptide:
- <200 pmol/L → Type 1 diabetes; start insulin therapy 2, 5
- 200-600 pmol/L → Consider MODY testing if clinical features present; otherwise treat as type 1 diabetes 5
600 pmol/L → Type 2 diabetes; consider oral agents or GLP-1 agonists 5
Critical Pitfalls to Avoid
Misdiagnosis occurs in 40% of adults with new type 1 diabetes, often misclassified as type 2 diabetes due to obesity or older age. 2, 5
Recent data show 24% of children with type 1 diabetes are overweight and 15% are obese—do not exclude type 1 diabetes based on body habitus alone. 1
In insulin-treated patients, C-peptide must be measured prior to insulin discontinuation to exclude severe insulin deficiency. 2, 3
C-peptide testing is only indicated in insulin-treated patients for classification purposes; in non-insulin-treated patients, clinical features and autoantibodies should guide diagnosis. 5
Ketosis-prone type 2 diabetes can present with DKA but maintains higher C-peptide levels (>600 pmol/L), distinguishing it from type 1 diabetes. 5
Special Populations
Children and Adolescents
More than half of children with newly diagnosed diabetes have residual beta-cell function (C-peptide >0.2 nmol/L) at diagnosis. 7
Only 1 in 1037 children with C-peptide <0.2 nmol/L at diagnosis had a diabetes type other than type 1. 7
Monogenic diabetes (MODY) accounts for 1.2-4% of pediatric diabetes and is frequently misdiagnosed as type 1 diabetes. 1