ECE2019 Poster Presentations Diabetes, Obesity and Metabolism 1 (104 abstracts)
1Hormonology Department, Cochin Hospital, Paris, France; 2Inserm U1016-CNRS UMR8104-Université Paris Descartes, Cochin Institute, Paris, France; 3Automatized biological diagnosis department, Cochin Hospital, Paris, France; 4Endocrinology Department, Cochin Hospital, Paris, France; 5INSERM U1139, Université Paris Descartes, Paris, France.
Introduction: Oral glucose tolerance test (OGTT) allows classification of subjects in 3 groups, depending on glycaemia 120 minutes after 75 g glucose ingestion: normal (glycaemia < 1.4 g/L), glucose intolerant (1.42 g/L) and diabetic (>2 g/L). Five insulin profiles associated with different incidence rates of diabetes over 10 years of follow-up have also previously been described. Insulin measurement is very sensible to hemolysis and can advantageously be replaced by C-peptide determination. However, little is known about C-peptide response to OGTT.
Material and methods: 128 patients were included to evaluate glyceamia (COBASE801® ROCHE Diagnostics, France), insulin and C-peptide (LiaisonXL®, Diasorin, France) responses to OGTT.
Results: According to Hitashi classification, 23 (18%) patients of the whole cohort harbored a physiological insulin response corresponding to profile I (peak of insulin during OGTT at 30 min and higher insulin level at 60 vs 120 min), 14 (11%) patients were classified in profile II (peak of insulin at 30 min and lower or equal insulin level at 60 vs 120 min), 56 (44%) in profile III (peak of insulin at 60 min), 26 (20%) in profile IV (peak of insulin at 120 min and lower insulin level at 30 vs 60 min), and finally 9 (7%) in profile V (peak of insulin at 120 min and higher or equal insulin level at 30 vs 60 min). Only 4 different mean C-peptide profiles emerged from the 5 subgroups previously defined by insulin profile, mean C-peptide profile being globally similar to mean insulin profile. The only difference relied on a similar C-peptide profile corresponding to a growing curve from T0 to T120 in both patients with insulin profile IV and V. Mean and 95% confidence interval of C-peptide value at the different times of OGTT were also calculated in the subgroup of patients with both normal glycemic and insulin (pattern I) responses to propose reference values: respectively T0: 0.53 (0.260.77); T30: 2.2 (1.243.29); T60: 2.26 (1.363.68); T120: 1.88 (0.842.62) nmol/L.
Conclusion: C-peptide response to OGTT profile seems to give globally the same information as insulin profile and should therefore also be predictive of the risk type 2 diabetes. The slight differences observed between the 2 parameters can be explained by different metabolism kinetics. This work also allows us to propose for the first time reference values for C-peptide at the different times of OGTT using Liaison XL® (Diasorin, France).