Key Takeaways
- C-peptide has a half-life of roughly 30 to 35 minutes versus 3 to 5 minutes for insulin, making it a more stable and less variable marker of beta-cell output in a single blood draw.
- C-peptide is the only reliable test to assess residual insulin secretion in a patient already injecting insulin, because exogenous insulin contains no c-peptide.
- In factitious hypoglycemia from covert insulin injection, serum insulin is elevated while c-peptide is suppressed, a dissociation that insulin testing alone cannot detect.
- Renal impairment (eGFR below roughly 30 mL/min/1.73m2) falsely elevates c-peptide by reducing clearance, a limitation commodity lab explainers almost never mention.
- Neither test alone diagnoses insulin resistance or PCOS; HOMA-IR from fasting glucose plus fasting insulin is the standard research surrogate, still without a universally agreed clinical cutoff.
The 50-Word Answer
C-peptide vs insulin level: c-peptide is the better test for assessing beta-cell function and residual secretion in people on insulin therapy, because exogenous insulin does not raise c-peptide and c-peptide has a much longer half-life. Serum insulin is preferred when investigating covert insulin injection or when real-time secretion kinetics are needed.Table of Contents
- What each test actually measures
- The biochemistry with specific numbers
- Evidence ledger: what each claim is built on
- When clinicians order one vs the other
- Honest head-to-head comparison table
- What most pages get wrong: renal clearance and assay interference
- The chemistry behind the rules of thumb
- How to read your own results: reference ranges and red flags
- C-peptide and insulin in insulin resistance workups
- FAQ
- Sources
What Does Each Test Actually Measure?
Both tests come from the same molecule. The pancreatic beta cell first synthesizes proinsulin, a single 86-amino-acid chain. Before secretion, a cleavage enzyme cuts out the middle segment, called connecting peptide or c-peptide (31 amino acids), releasing it alongside insulin in equimolar amounts. This means one molecule of c-peptide is made for every one molecule of insulin, every time a beta cell fires.
Check your GLP-1 eligibility
Use our free BMI Calculator to see if you may qualify for provider-reviewed GLP-1 therapy.
Try the BMI Calculator →Serum insulin measures circulating insulin, both what the beta cell just secreted and, critically, any insulin injected from outside. C-peptide measures only the connecting peptide produced endogenously. Because c-peptide is not part of any pharmaceutical insulin formulation, it cannot be injected to fake a result.
The liver extracts roughly 50 percent of portal insulin on the first pass, so peripheral serum insulin reflects post-hepatic levels. C-peptide is not extracted by the liver on first pass and is cleared primarily by the kidneys. This differential clearance is why peripheral c-peptide levels are approximately two to five times higher than equimolar insulin on a molar basis, and why c-peptide is more stable across a blood draw window.
The Biochemistry With Specific Numbers
Half-life is the single most important number for understanding why these two tests behave differently.
- Insulin half-life in plasma: approximately 3 to 5 minutes (established in multiple tracer studies; Polonsky KS et al., Journal of Clinical Investigation, 1988, used deconvolution modeling to characterize insulin secretion and clearance rates).
- C-peptide half-life in plasma: approximately 30 to 35 minutes (Polonsky and Rubenstein's foundational work on c-peptide kinetics established this range; it is roughly six to ten times longer than insulin).
- Renal clearance of c-peptide: the kidney accounts for the dominant route of c-peptide degradation and excretion. This is why urine c-peptide creatinine ratio (UCPCR) is a validated clinical tool (Besser REJ et al., Clinical Chemistry, 2011).
- Hepatic extraction of insulin: roughly 40 to 60 percent per portal pass, meaning peripheral insulin levels systematically underrepresent portal (beta-cell-secreted) insulin.
What these numbers do NOT prove: a longer half-life does not mean c-peptide is always clinically superior. In situations where you need to detect the rapid, minute-to-minute insulin swings of an insulinoma during a provocative test, the sharper kinetics of insulin may actually carry more diagnostic information. The longer half-life is an advantage for a single fasting draw, not necessarily for dynamic testing.
Evidence Ledger: What Each Claim Is Built On
| Claim | Best Evidence Type | Effect Direction | Confidence |
|---|---|---|---|
| C-peptide half-life is roughly 6 to 10x longer than insulin half-life | Human pharmacokinetic studies (Polonsky, Rubenstein et al.) | Established difference | High |
| C-peptide unaffected by exogenous insulin injection | Clinical chemistry principle, confirmed in factitious hypoglycemia case series | Confirmed absence of cross-reactivity | High |
| Suppressed c-peptide with elevated insulin distinguishes factitious hypoglycemia | Human diagnostic cohort studies and case series | Highly specific pattern | High |
| 72-hour fast criteria (insulin above 3 uIU/mL, c-peptide above 0.2 nmol/L) diagnose insulinoma | Human prospective series; Endocrine Society Guideline (Service FJ et al.) | Established diagnostic threshold | High |
| Renal impairment elevates c-peptide independently of beta-cell function | Human cross-sectional studies in CKD populations | C-peptide rises with falling eGFR | High |
| UCPCR reliably classifies diabetes type and residual secretion | Human validation study (Besser et al., 2011, n=179 across diabetes types) | Strong correlation with stimulated c-peptide | High |
| Fasting insulin as a surrogate for insulin resistance (HOMA-IR) | Population-level validation; no agreed clinical diagnostic cutoff | Positive association with IR markers | Moderate |
| C-peptide predicts cardiovascular risk independently | Epidemiological cohort studies; confounding by obesity/IR not fully resolved | Directional association | Low to Moderate |
| Fasting insulin reference ranges below 10 uIU/mL as "optimal" | Functional medicine consensus; not validated against hard clinical endpoints | Not established as diagnostic | Low |
When Do Clinicians Order One vs the Other?
Order c-peptide when:
- Classifying diabetes type in a patient already on insulin therapy (type 1 vs type 2 vs MODY).
- Assessing residual beta-cell function in a known type 1 patient (e.g., in clinical trials of beta-cell preservation therapies).
- Evaluating hypoglycemia in a patient who may be injecting insulin covertly.
- Monitoring beta-cell function after pancreas transplant or islet cell transplant.
- Classifying MODY subtypes: a stimulated c-peptide above approximately 0.2 nmol/L with spontaneous hyperglycemia suggests non-type-1 etiology per published MODY diagnostic pathways (Shields et al., Diabetic Medicine, 2012).
Order insulin level when:
- Investigating hypoglycemia in a patient not on insulin therapy, where a high insulin confirms endogenous or exogenous hyperinsulinism before c-peptide is checked.
- Calculating HOMA-IR (fasting glucose x fasting insulin / 405 using conventional units) for insulin resistance research or clinical screening.
- During dynamic tests (oral glucose tolerance, mixed meal tolerance) where the shape of the insulin curve carries diagnostic information.
- Assessing for sulfonylurea-driven hypoglycemia: elevated insulin and elevated c-peptide with normal or elevated proinsulin, combined with urine sulfonylurea screen.
Honest Head-to-Head Comparison
| Feature | C-Peptide | Serum Insulin | Winner (or Draw) |
|---|---|---|---|
| Reflects beta-cell secretion in insulin-treated patients | Yes, unaffected by exogenous insulin | No, contaminated by injected insulin | C-peptide |
| Detects covert insulin injection | Suppressed (low), helps confirm | Elevated, raises suspicion | Both required together |
| Stability for a single fasting draw | More stable; 30 to 35 min half-life | Less stable; 3 to 5 min half-life | C-peptide |
| Sensitivity to rapid secretory pulses | Smoothed out by longer half-life | Better reflects acute spikes | Insulin |
| Reliability in chronic kidney disease | Falsely elevated (reduced renal clearance) | Less kidney-dependent clearance | Insulin (in advanced CKD) |
| Use in insulinoma diagnosis | Required (confirms endogenous source) | Required (confirms hyperinsulinism) | Both required per guideline |
| HOMA-IR calculation | Not used in standard formula | Standard component of HOMA-IR | Insulin |
| Validated urine spot test available | Yes, UCPCR validated clinically | No validated urine form | C-peptide |
| Assay interference from insulin analogs | Not affected | Variable cross-reactivity by assay and analog | C-peptide |
What Most Pages Get Wrong: Renal Clearance and Assay Interference
In patients with eGFR below roughly 30 mL/min/1.73m2, c-peptide can be substantially elevated even in a patient with severely reduced beta-cell function. This is not a trivial edge case. Type 2 diabetes is the leading cause of chronic kidney disease. A clinician ordering c-peptide in a patient with both diabetes and advanced CKD without also checking creatinine and eGFR can badly misread the result.
The second omission is insulin assay interference. Standard insulin immunoassays vary significantly in their cross-reactivity with insulin analogs (glargine, lispro, aspart, detemir). Some older assays cross-react substantially with insulin glargine, meaning a patient on basal insulin therapy can appear to have endogenous hyperinsulinism on an insulin level if the laboratory's assay is not validated against that specific analog. C-peptide has no such analog interference problem.
A third omission is the proinsulin ratio. In insulinoma, the ratio of proinsulin to total insulin is typically elevated (above roughly 25 percent by some criteria), a refinement that requires ordering proinsulin separately and that neither a c-peptide nor an insulin level alone provides. Commodity pages discussing insulinoma diagnosis often present c-peptide and insulin as sufficient when current guidelines (Endocrine Society, 2009) also recommend proinsulin.
The Chemistry Behind the Rules of Thumb
Why does c-peptide have a longer half-life? It is not about molecular size. C-peptide (31 amino acids, roughly 3,020 Da) is smaller than mature insulin (51 amino acids, roughly 5,808 Da as a monomer). The half-life difference is explained by differential clearance pathways. Insulin binds insulin receptors throughout the body, and receptor-mediated endocytosis followed by intracellular degradation is a major clearance mechanism. C-peptide does not bind insulin receptors. It was thought for decades to be biologically inert. Its clearance depends almost entirely on renal filtration and tubular degradation. The kidney filters c-peptide efficiently, but with a finite capacity, so when glomerular filtration falls, c-peptide accumulates.
Why does hepatic first-pass extraction matter? The portal vein delivers insulin secreted by beta cells directly to the liver before it enters systemic circulation. The liver's insulin receptors extract a substantial fraction (estimates range from 40 to 60 percent per pass). C-peptide bypasses this extraction because hepatocytes do not significantly metabolize it. This is why a peripheral blood draw for c-peptide more accurately reflects total beta-cell output than a peripheral insulin draw does.
Why does exogenous insulin suppress c-peptide? Injected insulin enters systemic circulation and activates insulin signaling in the hypothalamus and through peripheral glucose-lowering. As blood glucose falls, the normal feedback loop suppresses beta-cell secretion. Fewer beta cells firing means less proinsulin cleaved, less c-peptide released. So in covert insulin injection, you see the paradox: high insulin (from the injection), low glucose (from insulin's action), and low c-peptide (because beta cells are appropriately suppressed by the falling glucose). This dissociation is the diagnostic key.
How to Read Your Own Results: Reference Ranges and Red Flags
Reference ranges are assay-specific. Do not compare a result from one laboratory against the published range of a different laboratory. That said, commonly cited reference values in the endocrinology literature are:
| Test | Fasting Reference Range (approximate) | Units | Important Caveats |
|---|---|---|---|
| C-peptide (fasting) | 0.5 to 2.0 ng/mL (or 0.17 to 0.66 nmol/L) | ng/mL or nmol/L | Elevated in CKD; reduced in type 1; must be interpreted with glucose |
| Serum insulin (fasting) | 2 to 25 uIU/mL (varies widely by lab) | uIU/mL or pmol/L | Analog cross-reactivity varies; no agreed "optimal" upper limit |
| UCPCR (urine c-peptide:creatinine ratio) | Above 0.2 nmol/mmol suggests meaningful residual secretion | nmol/mmol | Must be from a post-meal or stimulated sample for maximum sensitivity |
| 72-hour fast: insulin threshold for insulinoma | Above 3 uIU/mL (by ultrasensitive assay) concurrent with glucose below 55 mg/dL | uIU/mL | Older criteria used higher cutoffs; confirm which assay generation was used |
Unit conversion note: C-peptide results in ng/mL can be converted to nmol/L by dividing by approximately 3.02 (reflecting molecular weight). Insulin in uIU/mL can be converted to pmol/L by multiplying by approximately 6.945, though the exact conversion factor varies slightly by manufacturer standard. Always use the unit your laboratory reports and compare against that laboratory's stated range.
Red flag patterns to recognize:
- Low c-peptide (below 0.6 ng/mL) with high glucose: beta-cell failure, check for type 1 or late type 2.
- High insulin with low or suppressed c-peptide: exogenous insulin use, covert or prescribed.
- High c-peptide with high insulin and low glucose: endogenous hyperinsulinism (insulinoma or sulfonylurea effect); check proinsulin and drug screen.
- High c-peptide with normal glucose and no insulin therapy: suspect insulin resistance compensatory secretion, or check eGFR before concluding anything.
C-Peptide and Insulin in Insulin Resistance Workups
This is the area where the most overclaiming happens in consumer health content. Neither fasting c-peptide nor fasting insulin has a validated, guideline-endorsed cutoff for diagnosing insulin resistance in clinical practice. What exists is HOMA-IR, calculated as fasting insulin (uIU/mL) multiplied by fasting glucose (mg/dL), divided by 405. HOMA-IR was derived by Matthews et al. in 1985 (Diabetologia) and has been validated as a research tool in large epidemiological studies. A HOMA-IR above 2.5 to 3.0 is widely cited as a threshold for insulin resistance in research literature, but this threshold is population-derived and varies with ethnicity, age, and the insulin assay used.
Fasting c-peptide has been proposed as an alternative or complement to fasting insulin for IR assessment, with the rationale that its longer half-life and freedom from first-pass extraction make it more reproducible. Some research groups have reported that fasting c-peptide correlates well with hyperinsulinemic-euglycemic clamp results (the gold standard for IR measurement). However, this has not translated into a clinical diagnostic standard with agreed cutoffs, and the renal clearance confounder remains a problem in the populations most likely to have insulin resistance (those with obesity and early CKD).
The honest clinical position: fasting insulin is the standard for HOMA-IR and is more widely validated for IR screening. C-peptide adds value when insulin results are confounded by exogenous insulin or analog cross-reactivity. In a purely metabolic health context with no exogenous insulin, the two tests are broadly interchangeable for IR assessment, and neither gives you a clean yes/no diagnosis.
Frequently Asked Questions
Sources
- Polonsky KS, Rubenstein AH. C-peptide as a measure of the secretion and hepatic extraction of insulin: pitfalls and limitations. Diabetes. 1984;33(5):486-494.
- Polonsky KS, Given BD, Van Cauter E. Twenty-four-hour profiles and pulsatile patterns of insulin secretion in normal and obese subjects. Journal of Clinical Investigation. 1988;81(2):442-448.
- Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28(7):412-419.
- Besser REJ, Shepherd MH, McDonald TJ, et al. Urine C-peptide creatinine ratio is a practical outpatient tool for identifying hepatocyte nuclear factor 1-alpha/hepatocyte nuclear factor 4-alpha maturity-onset diabetes of the young from long-duration type 1 diabetes. Diabetes Care. 2011;34(2):286-291.
- Service FJ, Natt N. The prolonged fast. Journal of Clinical Endocrinology and Metabolism. 2000;85(11):3973-3974.
- Cryer PE, Axelrod L, Grossman AB
Related peptide guides