Elsevier

The Breast

Volume 16, Issue 4, August 2007, Pages 429-435
The Breast

ORIGINAL ARTICLE
The use of a breast symmetry index for objective evaluation of breast cosmesis

https://doi.org/10.1016/j.breast.2007.01.013Get rights and content

Summary

The cosmetic result after breast surgery is an important marker in clinical studies. Most authors used subjective scales to judge breast cosmesis. However, inter-observer discrepancies are very high and the use of such subjective scales for prospective trials is highly disputed. In this study we present for the first time a new invented breast symmetry index (BSI©). This BSI is calculated by subtracting the size and the shape between both breasts (frontal view and side view). The BSI is measured with a software system called breast analysing tool (BAT©) from digital photographs. The photographs of 27 patients have been analysed with this software by different physicians to evolve inter-observer reproducibility. The Harris scale for subjective cosmetic analyses has been correlated with the BSI. In our study the inter-observer reproducibility was excellent (Pearson correlation r=0.9; p<0.05) and the BSI was able to significantly differentiate between good and bad cosmesis (BSI values from 0%d to 30%d is good, BSI>30%d is bad cosmesis). Thus the BSI may be used for clinical studies.

Introduction

Breast conserving therapy (BCT) resulted in improved quality of life and self-esteem of women undergoing surgical treatment due to breast cancer.1 The assessment of cosmetic outcome in breast surgery is especially pertinent, because patient satisfaction, beside oncologic outcome, is the predominant factor in determining quality-of-life. Until now, none of the prospective randomised BCT-trials were able to assess cosmetic outcome due to the lack of reproducible, accurate and user-friendly scales.

To interpret cosmetic outcome, we have to differentiate between subjective and objective measurements. Concerning subjective analyses we further have to compare between the patient's and the doctor's view as physicians usually judge cosmetic results different.2 Moreover, physicians have to be divided into experts and non-experts.3

Subjective analyses by patients are poorly reproducible. However, this variable is the most important to evaluate as the main goal is to improve patients’ quality-of-life. In a recent review by Ching et al.4 the authors identified body-image and quality-of-life measures to be of the greatest value in determining surgical outcomes. This meta-analysis assessed different questionnaires and self-assessment tests. These methods seem to be very reliable, however, conducting repeated measurements and interpreting results may be difficult. Moreover the method is very time consuming and may depend on the patients’ daily state of mind and age.

Regarding subjective analyses by physicians, various scales have been used to assess body image. Unfortunately, none of these methods achieved widespread use. Recently, Cardoso et al. demonstrated that subjective analyses by different experts working in different geographical areas are only fairly reproducible.5 Thus, subjective methods seem to be inappropriate for scientifically based clinical trials to analyse cosmetic outcome after BCT.

Concerning objective analyses, Pezner et al.6 invented the breast retraction index by measuring nipple location comparing both sides with each other. They demonstrated good correlation with subjective votes of breast symmetry, however, this index only compared nipple asymmetry. Thus, skin retractions by BCT may be missed by this scoring system. Van Limbergen et al. combined nipple retraction with breast shape by additional measuring distances to the breast borders. The author demonstrated good correlation with a subjective cosmetic score, however, this method is not able to evaluate all breast deformities, especially those in the upper quadrants thus underestimating breast asymmetry.7, 8 In this regard objective tools still need to be improved to be used for clinical trials.

The aim of the study was to invent and test an improved objective method to evaluate breast cosmesis (breast symmetry index=BSI). This method should be reproducible, easy to handle and significantly separate good from bad cosmetic results by a scale. Thus, we determined the usefulness of the BSI for a prospective randomised trial by comparing the BSI with subjective analyses conducted by patients, experts and non-experts.

Section snippets

Patients and methods

The basic principle of our cosmetic index, the breast symmetry index (BSI), may be explained as follows: If the operated breast does not differ in size and shape from the contra lateral side, the symmetry is perfect and cosmetic outcome is good. Breast symmetry can easily be mathematically measured by comparing the area and the circumference as well as the nipple position of both breasts and subtracting the data of one breast from the contra lateral side. Therefore, digital pictures (side and

BAT software accuracy

Accuracy of the BAT software to measure distances from digital photos is demonstrated in Fig. 3 showing significant correlation between hand-measured distances and BAT-measured distances of from “dummies”. Both measurements correlate nearly 100% (r=0.998; p<0.05).

Clinical accuracy

In patients the BSI %difference [%d] indices ranged from 5 to 126%d with a median of 20 for the frontal %d, 13 for the side %d and 16 for the total %d. The BSI difference factor [df] ranged from 1.10 to 4.38 with a mean of 1.26, 1.29

Discussion

There is a lack of evidence based on prospective randomised trials regarding factors to improve breast cosmesis and patients’ quality-of-life after breast surgery. Several retrospective analyses demonstrated that an increased amount of excised breast tissue as well as boost radiation and tumour size correlated with a poor cosmetic outcome.10 Panel evaluation from different experts have been proven to be the most reliable and accurate way to evaluate cosmesis.3, 10 However, this method is

Acknowledgement

We would like to thank our patients for their dedicated help. Moreover, we would like to thank Mag. Karl Thomanek for manuscript preparations and “mammamia” for financial support.

Cited by (0)

View full text