Estimation of adipose tissue mass by magnetic resonance imaging: validation against dissection in human cadavers.

The evaluation of adipose tissue distribution has become an essential component of investigations on the complications of obesity. However, a major limitation is lack of methodology for accurate estimation of adipose tissue mass in the different regions of the body. Therefore, we have tested the accuracy and precision of magnetic resonance imaging (MRI) as a method to measure adipose tissue mass in regions of the body not accessible with standard anthropometric methods. The mass of subcutaneous and intraabdominal adipose tissue estimated by MRI was compared with that obtained by direct weighing of the same adipose tissue compartments after dissection in human cadavers. MRI was performed on three unembalmed cadavers (two males, one female) who were subsequently dissected to isolate intraperitoneal, retroperitoneal, and subcutaneous adipose tissues. These same components were delineated by MRI. The results of the two methods were highly congruent. For the various compartments, the mean of the difference between the two methods was only 0.076 kg (95% confidence interval + 0.005 kg and + 0.147 kg). The "limits of agreement" between the two techniques were -0.066 kg and +0.218 kg. Multiple repeated estimates of mass of adipose tissue compartments were made to determine reproducibility of the MRI measurement; the coefficient of variation for repeated measures was below 14%. The results of this study show that MRI is an accurate and precise technique to evaluate adipose tissue mass in subcutaneous and intraabdominal compartments. Furthermore, MRI was found to be a valid method to separately evaluate the mass of intraabdominal subcompartments of intraperitoneal and retroperitoneal adipose tissue.