This article waslast modified on 12 February 2018.

Equal amounts of insulin and C-peptide are produced in the pancreas when proinsulin is split, and they are stored and released into the blood together. Diabetic patients’ own insulin production can be assessed by measuring the serum concentration of C-peptide, even in those being treated with insulin. C-peptide can be used to help classify diabetes into early onset Type 1, in which little or no insulin is produced, and maturity onset Type 2 where the patients can produce insulin but either too little insulin for their needs or they have become resistant to it. Type 1 patients require treatment with insulin injections while Type 2 patients usually respond to dietary changes and tablets that promote the release of insulin. Sometimes rare genetic forms of Type 2 diabetes may be misdiagnosed as Type 1 and be inappropriately treated with insulin. In these patients, the response of serum C-peptide to food intake may indicate treatment by tablets instead of by insulin injections, and in Type 2 diabetics treated with tablets a failure of a C-peptide response may indicate the need for insulin injections. A drawback to the use of C-peptide in serum is that it is rapidly broken down, so patients have to attend hospital so that blood samples can be swiftly sent to the laboratory and frozen.

Professor Andrew Hattersley of the Clinical Research Facility at the Peninsula Medical School, Exeter and colleagues have previously shown that C-peptide is stable in urine preserved with boric acid at room temperature for up to 72 hours (Clinical Chemistry, November 2009). In two papers (Diabetes Care, online 1 February 2011, in print March 2011 and Diabetic Medicine accepted online 23 February 2011) the researchers have now compared the ratio of C-peptide to creatinine in urine with serum C-peptide after a meal in children and adults with Type 1 and Type 2 diabetes.

The gold standard test for showing that patients can produce their own insulin is to measure serum C-peptide after stimulation with a standard mixed meal. In the first investigation 72 patients with Type 1 diabetes who had been diagnosed at an average age of 14 had this test and also had urine collected to measure the ratio of C-peptide to creatinine. Urine for this measurement was also collected at home after an evening meal. There was good agreement between the serum and urine measurements. The researchers were able to define a cut off urine value that had 94% sensitivity and 100% specificity for the presence of significant insulin production. In the second investigation 51 patients with insulin-treated Type 2 diabetes diagnosed after the age of 30 had similar tests. The researchers were able to find a cut-off urine value that had 100% sensitivity and 96% specificity for the absence of significant insulin production.

The researchers concluded that the urine C-peptide creatinine ratio agrees well with serum C-peptide and may provide a practical alternative for detecting insulin production and insulin deficiency in clinical practice, avoiding the need for hospital admission.