Shoulder girdle impairment in breast cancer survivors: the role of range of motion as predictive factor for dose distribution and clinical outcome

Fabio Marazzi, Valeria Masiello, Domenico Marchesano, Luca Boldrini, Stefano Luzi, Paola E Ferrara, Eugenia Amabile, Danila Piccari, Francesco Landi, Francesca Moschella, Gianluca Franceschini, Riccardo Masetti, Giovanna Mantini, Vincenzo Valentini, Daniela Smaniotto

Research output: Contribution to journalArticlepeer-review


BACKGROUND: Pain and functional impairment of the ipsilateral shoulder girdle in patients who underwent surgery and radiotherapy for breast cancer (BC) is a late complication reported in the literature. We analyze a correlation with dosimetric parameters and propose an algorithm for sparing strategies.

METHODS: A total of 111 patients treated for BC were included in this observational analysis during follow-up protocol visits. Exclusion criteria were the presence of moderate or severe arthrosis history and/or rheumatologic diseases. All the patients had complete physical and multidimensional examinations during joint (physiatrist and radiotherapy oncology) follow-up visits. A scapula-humeral articulation (SHA) standardized contouring was performed retrospectively on Eclipse® treatment plans. A possible correlation between patients' characteristics, radiotherapy, and dosimetry analysis and functional impairment was investigated at statistical analysis. Results of analysis were summarized into a proposal of algorithm for sparing SHA.

RESULTS: A total of 111 patients were selected during follow-up visits. Mean age of patients was 60 years (range 41-85 years). A total of 103 patients (93%) underwent conservative surgery, with 110 patients (99%) undergoing axilla surgery as well. Fifty-two patients (46.8%) presented a reduction of range of motion (ROM) abduction on the treated side at the observational analysis. Mean ROM abduction reduction was 13°06' (range 0°-100°). Disability of the Arm, Shoulder and Hand questionnaire (DASH) score results were excellent in 79 patients (71.2%), discrete in 15 patients (13.5%), good in 15 patients (13.5%), and sufficient in 2 patients (1.8%). Median EQD2 Dmax at SHA was 18 Gy (range 0.22-51.9 Gy) and median EQD2 mean dose at SHA was 2 Gy (range 0.04-24.32 Gy). Univariate analysis showed a linear correlation between DASH score and ROM of abduction of treated side (ρ=-0.7), ROM of abduction and ROM of flexion in ipsilateral arm (ρ=0.8), or ROM of abduction and ROM of flexion in contralateral arm (ρ=0.8). A statistically significant difference in ROM abduction between the 2 arms was found at χ2 test (P<0.05 at χ2 confidence interval = 95%). Cox linear regression analysis showed ROM abduction on treated arm as a predictive factor of DASH score (P<0.0001). Age (P<0.05), DASH score (P=0.006), and ROM abduction on treated arm (P=0.005) were found as independent predictive factors of mean dose at multivariate analysis. A mean dose higher than 7 Gy and ROM abduction reduction more than 30° were related to DASH score level reduction.

CONCLUSIONS: This hypothesis-generating study introduces an algorithm to be validated for management of sparing SHA and improving quality of survivorship. ROM evaluation after surgery, early physiotherapy, standard contouring, and planning adaptation represent possible indications to preserve shoulder impairment. Further prospective studies are needed to discriminate impairment of surgery and radiotherapy in order to personalized therapeutic plan programs.

Original languageEnglish
Pages (from-to)319-330
Number of pages12
Issue number4
Publication statusPublished - Aug 2019


  • Adult
  • Aged
  • Aged, 80 and over
  • Breast Neoplasms/physiopathology
  • Cancer Survivors
  • Cross-Sectional Studies
  • Female
  • Humans
  • Middle Aged
  • Range of Motion, Articular/physiology
  • Retrospective Studies
  • Scapula/physiopathology
  • Shoulder/physiopathology
  • Shoulder Joint/physiopathology


Dive into the research topics of 'Shoulder girdle impairment in breast cancer survivors: the role of range of motion as predictive factor for dose distribution and clinical outcome'. Together they form a unique fingerprint.

Cite this