Long-term Effect of C-Peptide Level on Clinical Outcomes in the Scottish Type 1 Bioresource Cohort.

Abstract

To characterize associations of C-peptide measurements with type 1 diabetes (T1D) complications across long-term clinical follow-up in a large representative population cohort (Scottish Diabetes Research Network Type 1 Bioresource [SDRNT1BIO]). SDRNT1BIO includes people with a clinical diagnosis of T1D who are aged 16 years or older at enrollment. Their median diabetes duration is 21 (interquartile range = 11-31) years. C-peptide was measured in an untimed blood sample using the Roche immunoassay with a lower limit of detection of 3 pmol/L. Incident events and continuous outcomes were captured through electronic health record linkage across follow-up. Associations were studied using Poisson generalized linear models and linear mixed models adjusted for sex, age, time-updated T1D duration, year of entry, and age/T1D duration interaction. Participants (n = 5,630) were followed for a median of 10.8 years. Inverse associations were found between baseline C-peptide level and diabetic ketoacidosis (DKA) (P < 0.0001; no. of events = 1,015), severe hospitalized hypoglycemia (SHH) (P = 0.0055; n = 539), and incident retinopathy (P < 0.0001; n = 1,172). These associations persisted when adjusted for time-updated glycosylated hemoglobin. No associations were found between baseline C-peptide levels and cardiovascular disease (P = 0.9; n = 714) or death (P = 0.8; n = 566). In a subset of participants (n = 407) with serial C-peptide measures, follow-up C-peptide level was inversely associated with risk for DKA (P < 0.001; n = 92) and there was a borderline association with SHH (P = 0.015; n = 28). These data demonstrate the importance of C-peptide level and its maintenance for glycemic control and avoidance of acute and chronic complications of T1D.