C peptide assay

Potential mediators comprised low physical activity level, estimates of hyperinsulinaemia/insulin resistance and glycaemic control, lipids, blood pressure, and comorbidities. Physical activity was self-reported using the validated Saltin-Grimby Physical Activity Level Scale. 20 (link) Data on glycated haemoglobin (HbA1c), triglycerides, low-density lipoprotein (LDL) cholesterol, and systolic and diastolic blood pressure were obtained from the Danish Diabetes Database of Adults (DDDA) using the hospitals' routine analysis procedures closest to the date of enrolment. Fasting serum C-peptide (pmol/L), as a marker of both hyperinsulinaemia and insulin resistance, was analysed using the Roche C-peptide assay (Roche Diagnostics, Mannheim, Germany). Fasting derived indices of insulin resistance and beta-cell function were estimated using homeostatic model assessment (HOMA2-IR and HOMA2-B, respectively) from fasting C-peptide and plasma glucose levels. 21 (link) Information on comorbidities according to the Charlson comorbidity index, excluding diabetes, was obtained via linkage to the Danish National Patient Registry tracked 10 years back from date of enrolment. 22 (link)

All biochemical tests using blood samples from the DD2 biobank were performed in the laboratory at Centre Hospital Lillebaelt, Region of Southern Denmark. The laboratory is accredited according to the ISO 15189 standard. Fasting C-peptide was analysed using the Roche C-Peptide assay (Roche Diagnostics, Mannheim, Germany) and fasting plasma glucose was analysed using an enzymatic hexokinase method (Glucoquant Glucose/HK, Roche Diagnostics, Mannheim, Germany). We used version 2 of the revised homeostatic assessment model (HOMA2) to estimate insulin sensitivity (HOMA2S) and beta cell function (HOMA2B) based on fasting plasma glucose values and C-peptide 24 (link) . HOMA2 parameters were categorized according to previously published thresholds 22 (link) . We analysed GAD-antibodies using the This article is protected by copyright. All rights reserved.
AESKULISA GAD65 kit (AESKU Diagnostics, Wendelsheim, Germany). Pancreas specific amylase was analysed for the first 996 consecutive patients enrolled in the DD2 cohort using the COBAS-6000 analyser produced by Roche Diagnostics GmbH, Mannheim, Germany. A cut-off <17 U/l was used to define patients with a low amylase level 25 (link) . The remaining biochemical parameters (triglycerides, HbA1c and urine albumin/creatinine ratio) were collected from linked medical and administrative registries 18 (link) .