Send to

Choose Destination
Int J Cosmet Sci. 2015 Feb;37(1):134-40. doi: 10.1111/ics.12177. Epub 2014 Nov 19.

Cellulite: poor correlation between instrumental methods and photograph evaluation for severity classification.

Author information

Department of Dermatology, Federal University of Sao Paulo, UNIFESP, Sao Paulo, Brazil.



Cellulite refers to skin relief alterations in women's thighs and buttocks, causing dissatisfaction and search for treatment. Its physiopathology is complex and not completely understood. Many therapeutic options have been reported with no scientific evidence about benefits. The majority of the studies are not controlled nor randomized; most efficacy endpoints are subjective, like not well-standardized photographs and investigator opinion. Objective measures could improve severity assessment. Our purpose was to correlate non-invasive instrumental measures and standardized clinical evaluation.


Twenty six women presenting cellulite on buttocks, aged from 25 to 41, were evaluated by: body mass index; standardized photography analysis (10-point severity and 5-point photonumeric scales) by five dermatologists; cutometry and high-frequency ultrasonography (dermal density and dermis/hypodermis interface length). Quality of life impact was assessed. Correlations between clinical and instrumental parameters were performed.


Good agreement among dermatologists and main investigator perceptions was detected. Positive correlations: body mass index and clinical scores; ultrasonographic measures. Negative correlation: cutometry and clinical scores. Quality of life score was correlated to dermal collagen density.


Cellulite caused impact in quality of life. Poor correlation between objective measures and clinical evaluation was detected. Cellulite severity assessment is a challenge, and objective parameters should be optimized for clinical trials.


cellulite; instrumental methods; statistics

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Wiley
Loading ...
Support Center