Quantification of the visceral and subcutaneous fat by computed tomography: interobserver correlation of a single...

To assess the interobserver reproducibility of the quantification of the visceral and subcutaneous fat by computed tomography from an umbilical slice and study the effect of the level of the slice (slice going through the navel versus a slice going through disc L3-L4).

Forty-four breast cancer patients who had a CT-scan were included in this study. This is a double blind (junior versus senior) retrospective study to determine the interobserver reproducibility. A junior observer studied the variation between two levels of slice by selecting an image going through L3-L4 and the navel.

The measurement of the fat obtained from an umbilical slice seemed to be well correlated and consistent with that obtained from a slice with a disc reference (L3-L4). The interobserver reproducibility is good for the quantification of the umbilical fat (Spearman and Lin at 0.9921 and 0.985 [P<0.001] for the visceral fat).

The interobserver reproducibility of the single slice CT-scan measurement going through the navel (easily detected) is excellent and may therefore be used in oncology as a predictive tool to measure a characteristic of the host and not the tumor.