Objective To investigate the prevalence and characteristics of shoulder pathologies in women with breast cancer–related lymphedema (BCRL) using magnetic resonance imaging (MRI), and to compare these findings with those in controls of comparable age distribution presenting with shoulder pain but without malignancy or lymphedema. Methods In this retrospective case‐control study, 45 women with BCRL and 45 controls of comparable age underwent shoulder MRI for unilateral shoulder pain. MRI findings included rotator cuff tendinopathies or tears, joint degeneration, and intra‐articular or bursal effusion. Associations between lymphedema‐related factors and effusion were examined using univariable and multivariable logistic regression analyses. Inter‐reader agreement for effusion detection was also assessed. Results MRI‐detected effusion was significantly more frequent in the BCRL group than in controls (73.3% vs. 15.6%, p < 0.001), while the frequency of normal shoulder findings was notably lower (2.2% vs. 17.7%, p = 0.03). Rotator cuff and biceps pathologies were more common in the BCRL group, but the differences did not reach statistical significance. Within the BCRL group, patients with effusion had a significantly longer mean lymphedema duration in descriptive comparisons ( p = 0.007), whereas lymphedema duration was not an independent predictor of effusion in logistic regression analysis. Lymphedema severity showed no significant association with specific MRI abnormalities. Conclusion Women with BCRL showed a higher prevalence of shoulder joint effusion on MRI than controls, while normal shoulder findings were less frequent. Effusion may represent a correlational imaging marker associated with chronic lymphatic dysfunction rather than a direct causal consequence. These findings support the selective use of imaging in BCRL patients presenting with shoulder complaints and highlight the need for prospective studies incorporating oncologic treatment parameters and functional outcomes.
Özçete et al. (Thu,) studied this question.