Body Composition Profiling using MRI - Normative Data for Subjects with Diabetes Extracted from the UK Biobank Imaging Cohort
2016 (English)Conference paper, Abstract (Other academic)
To describe the distribution of MRI derived body composition measurements in subjects with diabetes mellitus (DM) compared to subjects without diabetes.
METHOD AND MATERIALS
3900 subjects (1864 males and 2036 females) from the UK Biobank imaging study were included in the study. The age range was 45 to 78 years. Visceral adipose tissue volume normalized with height2 (VATi), total abdominal adipose tissue volume normalized with height2 (ATATi), total lean thigh muscle volume normalized with body weight (muscle ratio) and liver proton density fat fraction (PDFF) were measured with a 6 minutes 2-point Dixon imaging protocol covering neck to knee and a 10-point Dixon single axial slice protocol positioned within the liver using a 1.5T MR-scanner (Siemens, Germany). The MR-images were analyzed using AMRA® Profiler research (AMRA, Sweden). 194 subjects with clinically diagnosed DM (DM group) were age and gender matched to subjects without DM (control group). For each variable and group, the median, 25%-percentile and 75%-percentile was calculated. Mann-Whitney U test was used to test the observed differences.
VATi in the DM group was 2.13 (1.43-2.62) l/m2 (median, 25% - 75% percentile) compared to 1.32 (0.86 - 1.79) l/m2 in the control group. ATATi in the DM group was 4.94 (3.86-6.19) l/m2 compared to 3.40 (2.56 - 4.70) l/m2 in the control group. Muscle ratio in the DM group was 0.13 (0.11 - 0.14) l/kg compared to 0.14 (0.12 - 0.15) l/kg in the control group. Liver PDFF in the DM group was 7.23 (2.68 - 13.26) % compared to 2.49 (1.53 - 4.73) % in the control group. Mann-Whitney U test detected significant differences between the DM group and the control group for all variables (p<10-5).
DM is strongly associated with high visceral fat, liver fat, and total abdominal fat, and low muscle ratio.
Body composition profiling shows high potential to provide direct biomarkers to improve characterization and early diagnosis of DM.
Place, publisher, year, edition, pages
Radiology, Nuclear Medicine and Medical Imaging Medical Image Processing
IdentifiersURN: urn:nbn:se:liu:diva-132411OAI: oai:DiVA.org:liu-132411DiVA: diva2:1045469
Radiological Society of North America 2016, the 102nd Scientific Assembly and Annual Meeting, Chicago, Illinois, USA, November 27 - December 2, 2016