Can the Molar Insulin: C-Peptide Ratio Be Used to Predict Hyperinsulinaemia?

Abstract

Hyperinsulinaemia is the precursor to numerous metabolic disorders. Early diagnosis and intervention could improve population health. Diagnosing hyperinsulinaemia is problematic because insulin has a very short half-life (2-5minutes). It is theorised that c-peptide levels (half-life 20-30minutes) would be a better proxy for insulin due to both hormones being released in equimolar amounts. However, the correlation between c-peptide and insulin levels is unknown. We aim to identify their correlation following a four-hour oral glucose tolerance test (OGTT). Data were obtained from records of routine medical care at St Joseph's Hospital, Chicago, IL, USA, during 1977. Two hundred and fifty-five male and female participants aged over 20 years undertook a fourhour OGTT with plasma glucose, insulin and c-peptide levels recorded. Correlation was assessed with Pearson's correlation. There was a weak correlation between insulin and c-peptide, which increased to moderate across the four-hour OGTT (r = 0.482-0.680). There was no significant change in this relationship when data was subdivided according to either the WHO glucose status or Kraft insulin response. Although there was a correlation between insulin and c-peptide, it was too weak to recommend the use of c-peptide as an alternative biomarker for the diagnosis of hyperinsulinaemia.

Keywords: c-peptide; correlation; hyperinsulinaemia; insulin; oral glucose tolerance test; prediction.

Figure 1

Pearson correlation r values against time. Correlation strength can be interpreted as: Weak = 0.3–0.5, Moderate = 0.5–0.7, Strong = > 0.7.

Figure 2

Kraft insulin response pattern r values against time. Due to the limited number of participants, Kraft pattern IV and V were not used in further analysis.

Figure 3

Total area under the curve c-peptide against insulin over 240 min.

Figure 4

Confidence intervals for total AUC.

Figure 5

Insulin vs. c-peptide at 240 min.