[Landmark-based statistical procrustes analysis in the examination of breast shape and symmetry]

The aim of this presentation is the explanation of a mathematical analysis of a landmark-based procrustes calculation. The various components of breast symmetry and their impact on breast reconstruction have been examined.

In an objective breast shape analysis asymmetries are quantified as a measure of the difference between a 3D landmark configuration of a breast and its mirror image. 10 landmarks, 4 on each side and 2 for a definition of the midline were defined. For each landmark the 3D coordinates were calculated statistically with the help of the software Programme R. Over a midline from the jugulum to the manubrium sterni, a mirror image of the breast was constructed. The data for the breast to be reconstructed and the non-operated breast from 44 patients were investigated. The statistical differences of the Procrustes analysis were used for an asymmetry score and the proportions of the individual components were calculated. These included size, location and orientation of the breast as well as the individual inherent morphological surface form data of the breast.

All 44 patients exhibited breast asymmetries and the mean asymmetry score amounted to 0.52. A calculation of the proportions of the individual components making up the asymmetry score revealed that morphological surface data made the largest contribution to the asymmetry score, closely followed by location, i. e., positioning of the breast on the thoracic wall. In contrast, the size of the breast and its orientation were of lesser relevance.

When considering breast symmetry in plastic surgery, the form plays a more important role than the size as objectively calculated by the statistical procrustes analysis. Almost equally important as the form is the positioning of the breast on the thoracic wall which contributes significantly more to total breast symmetry than axial displacements.