A literature review of articles concerning fracture risk in Diabetes, and a population based follow up study on HbA1c values and fracture risk in “The Tromsø 4 health survey”.
Abstract In this study the first aim was to do a literary review on the relation between diabetes and fracture risk. Second aim is to study the association between Hba1c and fracture risk in the dataset from Tromsø 4. For the literature study PubMed was used, and search words were “fracture, risk and diabetes”. The search gave 370 articles, approximately 59 of these have been used. Data on fracture risk in diabetes and pathophysiological models are presented. The follow up study was based on the Tromsø 4 health survey, in which 7127 subjects participated. The fracture rates were investigated during a mean follow up time of 517 weeks. A COX-regression model, which was stratified for gender and adjusted for age, was used to study fracture risk in relation to HbA1c values. The COX- regression study also gave survival plots for both genders. A graphical presentation of fracture risk in the HbA1c-quartiles across all age groups, in both genders, was made to study the possible interaction of HbA1c*age and fracture risk. The text study showed consistency with regards to increased fracture risk in type 1 diabetes, in type 2 diabetes the numbers were not so conclusive, some studies even found a decreased fracture risk in t2DM. In diabetics with diabetic complications, fracture risk was increased in both diseases. Pathophysiological mechanisms in diabetic bone disease has been explored. In the follow up study there was a total of 799 fractures. The cumulative fracture incidence was 11% in the study population, 15 % females and 6 % in men during follow up. Fracture risk decreased with increasing HbA1c among women: by each unit increase in HbA1c fracture hazard ratio (HR) decrease by 14%, CI (-26-0%), (p=0,048). In men HR was decreased by 6%, CI (-25,5-18%), but the relation was insignificant at (p=0,611). The interaction of age*HbA1c and fracture risk was very weak in both genders and showed no significance at p=0,735 in females and 0,436 in males.
PublisherUniversitetet i Tromsø
The following license file are associated with this item: