Quantitative assessment of radiation-induced fibrosis of the breast with tissue compliance meter, palpation, and radiological imaging: preliminary results Academic Article uri icon

Overview

MeSH Major

  • Breast
  • Breast Neoplasms
  • Palpation
  • Radiation Injuries

abstract

  • Quantification of radiation (RT)-induced fibrosis (RIF) continues to present a challenge in breast cancer survivors. We compare assessment of RIF by palpation and tissue compliance meter (TCM) to the radiological findings in women treated with RT. Of 300 patients treated with adjuvant RT, 17 women had > or =2-year follow-up sufficient to document RIF. Palpation and TCM were employed by three radiation oncologists in a blinded fashion. Palpation grades 1, 2, and 3 denoted mild, moderate, and severe RIF. TCM measured degree of compliance (DC) of RIF in irradiated (RTB) and nonirradiated breasts (NRTB). Architectural distortions (AD) on mammograms, ultrasound (US), and MRI were assessed. Median time of follow up was 3.9 years (range 2.1-6.5 years). Palpation revealed RIF grades 1, 2, and 3 in four, 10, and three patients, respectively. Mean percent changes (PC) in DC between RTB and NRTB by TCM were 19.5%, 37.1%, and 57.5% for grades 1, 2, and 3 RIF, respectively (p < 0.0001). There was a strong linear correlation between palpation grade and PC of DC by TCM (spearman-rank correlation=0.88, p < 0.0001). Interobserver variability (reliability) was computed using intraclass correlation coefficient (ICC) for TCM and kappa statistic for clinical palpation (ICC=0.99 [p < 0.0001] and kappa=0.70 [p < 0.0001], respectively). There was no correlation between average size of the AD as measured by the imaging modalities and RIF as assessed by palpation or TCM. Our preliminary data suggest that quantification of RIF is best with TCM. TCM results correlate better with palpation than with radiological imaging. The study with larger number of patients required to confirm our findings is underway.

publication date

  • November 2009

Research

keywords

  • Academic Article

Identity

Language

  • eng

Digital Object Identifier (DOI)

  • 10.1111/j.1524-4741.2009.00835.x

PubMed ID

  • 19824999

Additional Document Info

start page

  • 583

end page

  • 92

volume

  • 15

number

  • 6