Optimal Timing for C-Peptide Measurement
For diabetes classification, obtain a random (non-fasting) C-peptide sample within 5 hours of eating, which is as reliable as formal stimulation testing and far more practical for routine clinical use. 1
Fasting vs. Random C-Peptide Testing
Random (Non-Fasting) C-Peptide – Preferred for Most Clinical Scenarios
A random C-peptide drawn within 5 hours of a meal can completely replace formal C-peptide stimulation tests for classifying diabetes type, making testing feasible whenever the patient presents to clinic. 1, 2
Random non-fasting C-peptide correlates extremely strongly (r = 0.91–0.96) with stimulated C-peptide from mixed meal tolerance tests, with sensitivity and specificity exceeding 90% for detecting clinically relevant thresholds. 3
The correlation improves further (r = 0.96) when the concurrent glucose is ≥8 mmol/L (≥144 mg/dL), so if glucose is <4 mmol/L (<70 mg/dL) at the time of sampling, consider repeating the test. 1, 3
Random C-peptide testing eliminates the logistical burden of scheduling fasting appointments or performing meal tolerance tests, dramatically improving clinical workflow without sacrificing diagnostic accuracy. 3
Fasting C-Peptide – Required Only for Insurance Documentation
Measure fasting C-peptide specifically when insurance payers require documentation for insulin pump therapy coverage, and only when the simultaneous fasting plasma glucose is ≤220 mg/dL (≤12.2 mmol/L). 1, 2
An 8-hour overnight fast is the standard protocol when fasting C-peptide is required. 2
Outside of insurance requirements, fasting C-peptide offers no clinical advantage over random testing for diabetes classification. 1
Critical Timing Restrictions
Never measure C-peptide within 2 weeks of a hyperglycemic emergency (diabetic ketoacidosis or hyperosmolar hyperglycemic state), as acute metabolic decompensation temporarily suppresses β-cell function and yields falsely low results. 1, 4, 2
In insulin-treated patients, C-peptide must be measured before discontinuing insulin to avoid precipitating severe hyperglycemia or ketoacidosis while attempting to assess endogenous insulin secretion. 1, 2
Interpretation Thresholds
C-peptide <200 pmol/L (<0.6 ng/mL) indicates type 1 diabetes with severe insulin deficiency requiring lifelong insulin therapy. 1, 4, 5
C-peptide 200–600 pmol/L (0.6–1.8 ng/mL) suggests type 1 diabetes, MODY, or long-standing insulin-treated type 2 diabetes; further testing with islet autoantibodies or genetic studies may be needed. 1, 2
C-peptide >600 pmol/L (>1.8 ng/mL) strongly indicates type 2 diabetes with preserved β-cell function, making the patient a candidate for oral agents and lifestyle modification rather than insulin. 1, 4
Special Clinical Scenarios
Hypoglycemia Evaluation
When investigating unexplained hypoglycemia to exclude surreptitious insulin use or diagnose insulinoma, C-peptide should be measured during a documented hypoglycemic episode (glucose <55 mg/dL), not at a scheduled time. 1, 4
For suspected insulinoma, a supervised 48–72 hour fast with simultaneous measurement of insulin, C-peptide, proinsulin, and glucose provides definitive biochemical confirmation. 4
Postprandial C-Peptide in Type 2 Diabetes
Postprandial C-peptide 2 hours after breakfast (CPR2h) is the single most reliable index for identifying non-obese type 2 diabetes patients who will require multiple daily insulin injections rather than oral agents alone. 6
This postprandial approach outperforms fasting C-peptide, glucagon-stimulated C-peptide, and other indices in predicting insulin requirement. 6
Common Pitfalls to Avoid
Do not use C-peptide for routine diabetes screening or monitoring in established diabetes—its role is limited to classification when diabetes type is ambiguous and to insurance documentation for pump therapy. 1
Do not delay testing to obtain a fasting sample unless specifically required for insurance; random testing is equally valid and far more convenient. 1, 3
Do not interpret C-peptide in isolation—always consider concurrent glucose, diabetes duration, presence of autoantibodies, and clinical phenotype. 1, 3
Very low C-peptide results (<80 pmol/L) do not require repeat testing, as they definitively establish severe insulin deficiency. 1