European Journal of Epidemiology

, Volume 26, Issue 1, pp 29–38

Estimation of the contribution of biomarkers of different metabolic pathways to risk of type 2 diabetes

Authors

    • Department of EpidemiologyGerman Institute of Human Nutrition Potsdam-Rehbruecke
  • Dagmar Drogan
    • Department of EpidemiologyGerman Institute of Human Nutrition Potsdam-Rehbruecke
  • Hans-Georg Joost
    • Department of PharmacologyGerman Institute of Human Nutrition Potsdam-Rehbruecke
  • Heiner Boeing
    • Department of EpidemiologyGerman Institute of Human Nutrition Potsdam-Rehbruecke
  • Andreas Fritsche
    • Department of Internal Medicine IVUniversity of Tuebingen
  • Erwin Schleicher
    • Department of Internal Medicine IVUniversity of Tuebingen
  • Matthias B. Schulze
    • Department of Molecular EpidemiologyGerman Institute of Human Nutrition Potsdam-Rehbruecke
  • Tobias Pischon
    • Department of EpidemiologyGerman Institute of Human Nutrition Potsdam-Rehbruecke
DIABETES

DOI: 10.1007/s10654-010-9539-0

Cite this article as:
Montonen, J., Drogan, D., Joost, H. et al. Eur J Epidemiol (2011) 26: 29. doi:10.1007/s10654-010-9539-0

Abstract

The contribution of different biological pathways to the development of type 2 diabetes was quantified in a case-cohort design based on circulating blood biomarkers from participants aged 35–65 years in the EPIC–Potsdam Study. The analytic sample included 613 participants with incident diabetes and 1965 participants without diabetes. The proportion that each biomarker contributed to the risk of diabetes was quantified using effect decomposition method. Summarized risk of each biomarker was estimated by an index based on quintiles of gamma-glutamyltransferase (GGT), HDL-cholesterol, hs-CRP, and adiponectin. Cox proportional hazard regression was used to estimate relative risks adjusted for age, sex, body mass index, waist-circumference, education, sport activity, cycling, occupational activity, smoking, alcohol intake, and consumptions of red meat, coffee and whole grain bread. Adiponectin explained a total of 32.1% (CI = 16.8, 49.1%) of the risk related to index. For the other biomarkers the corresponding proportions were 23.5% (CI = 10.1, 37.8%) by HDL-cholesterol, 21.5% (CI = 11.5, 32.8%) by GGT, and 15.5% (CI = 4.44, 27.3%) by hs-CRP. The results support the hypothesis that the different biological pathways reflected by GGT, HDL-cholesterol, hs-CRP and adiponectin independent from each other contribute to the risk of type 2 diabetes. Of these pathways the highest contribution was observed for adiponectin which contributed one-third to the risk and that equal proportion was contributed by GGT and HDL-cholesterol, although the contribution of inflammation was lower.

Keywords

Adiponectin/blood Biological markers/blood Diabetes Liver enzymes Inflammation Prospective studies

Copyright information

© Springer Science+Business Media B.V. 2010