[Rapid total body fat measurement by magnetic resonance imaging: quantification and topography].

PURPOSE

To evaluate a rapid and comprehensive MR protocol based on a T1-weighted sequence in conjunction with a rolling table platform for the quantification of total body fat.

MATERIALS AND METHODS

11 healthy volunteers and 50 patients were included in the study. MR data was acquired on a 1.5-T system (Siemens Magnetom Sonata). An axial T1-weighted flash 2D sequence (TR 101, TE 4.7, FA 70, FOV 50 cm, 205 x 256 matrix, slice thickness: 10 mm, 10 mm interslice gap) was used for data acquisition. Patients were placed in a supine position on a rolling table platform capable of acquiring multiple consecutive data sets by pulling the patient through the isocenter of the magnet. Data sets extending from the upper to lower extremities were collected. The images were analyzed with respect to the amount of intraabdominal, subcutaneous and total abdominal fat by semi-automated image segmentation software that employs a contour-following algorithm.

RESULTS

The obtained MR images were able to be evaluated for all volunteers and patients. Excellent correlation was found between whole body MRI results in volunteers with DEXA (r (2) = 0.95) and bioimpedance (r (2) = 0.89) measurements, while the correlation coefficient was 0.66 between MRI and BMI, indicating only moderate reliability of the BMI method. Variations in patients with respect to the amount of total, subcutaneous, and intraabdominal adipose tissue was not related to standard anthropometric measurements and metabolic lipid profiles (r (2) = 0,001 to 0.48). The results showed that there was a significant variation in intraabdominal adipose tissue which could not be predicted from the total body fat (r (2) = 0.14) or subcutaneous adipose tissue (r (2) = 0.04). Although no significant differences in BMI could be found between females and males (p = 0.26), females showed significantly higher total and subcutaneous abdominal adipose tissue (p < 0.05).

CONCLUSION

This MR protocol can be used for the rapid and non-invasive quantification of body fat. The missing relationship between serum lipids and body fat masses suggests that the latter is an additional and independent hazard factor. Variations in body fat distribution, e. g. relationship between subcutaneous and intraabdominal fat, can be comprehensively assessed.