Send to

Choose Destination
See comment in PubMed Commons below
Coll Antropol. 2008 Jun;32(2):607-14.

Subcutaneous fat patterns in type-2 diabetic men and healthy controls.

Author information

Institute of Physiological Chemistry, Center of Physiological Medicine, Medical University Graz, Graz, Austria.


The optical device LIPOMETER enables the non-invasive, quick, and save determination of the thickness of subcutaneous adipose tissue layers at any given site of the human body. The specification of 15 evenly distributed body sites allows the precise measurement of subcutaneous body fat distribution, so-called subcutaneous adipose tissue topography (SAT-Top). In the present paper we focus on SAT-Top of male type-2 diabetes patients (N=21), describing very precisely their special SAT development and their SAT-Top deviation from a healthy control group (N=111), applying factor analysis and ROC curves. Factor analysis revealed three independent subcutaneous body fat compartments, which can be summarised as "upper body", "lower trunks" and "legs". The upper body SAT-Top is much more pronounced in diabetic men compared to their healthy controls (p<0.001). Furthermore, high diagnostic power by ROC curve analysis was achieved by different measurement sites of the upper body and summary measures of upper body obesity (sum2, which is the sum of neck and biceps, provides: area index =0.86, sensitivity =81%, specificity =90.1%, at an optimal cutoff value of 18.8 mm), ascribing a higher diabetes probability to subjects with a more upper body SAT-Top pattern. Calculating new ROC curves for diabetic patients with HBA1C values >8 (N=17) and their healthy controls (N=111) we received improved discrimination power for several SAT-Top body sites, especially for sum2, showing an area index of 0.91, a sensitivity of 94.1%, and a specificity of 90.1% at the optimal cutoff value of 18.8 mm. Concluding, the exact and complete description of the especial type 2 diabetic SAT pattern, which differs strongly from the SAT-Top of healthy controls, suggests the LIPOMETER technique combined with advanced statistical methods such as factor analysis and ROC curve analysis as a possible detecting tool for this disease.

[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Loading ...
    Support Center