Reply to “Letter to the Editor: Risk of Lymphedema After Sentinel Node Biopsy in Patients With Breast Cancer”

Dear Editors,

We appreciate the interest and valuable comments from Dr. Bergmann and Dr. Andrade [1] on our study “Risk of Lymphedema After Sentinel Node Biopsy in Patients With Breast Cancer” [2]. Their thoughtful feedback provided the opportunity to clarify several important aspects of our research.

To address their concerns regarding the incidence of lymphedema after radiation therapy, we performed additional analyses of radiation treatment sites. Among patients with lymphedema, significantly higher proportions received breast radiotherapy (RT, 77.6% vs. 55.4%, p = 0.003) and supraclavicular lymph node or internal mammary lymph node RT (57.1% vs. 14.1%, p < 0.001) compared to those without lymphedema. When all cases were considered, including the axillary region, the difference became significant (53.1% vs. 15.1%, p < 0.001). The overall rate of receiving RT was also significantly higher in the lymphedema group (79.6% vs. 55.5%, p = 0.001). However, it should be noted that these statistical comparisons should be interpreted with caution, because patients undergoing regional irradiation often receive radiation to multiple sites including the axilla and regional lymph nodes. This overlap between radiation sites makes it challenging to isolate the independent effect of each radiation site on the risk of subsequent lymphedema development.

Regarding the administration of chemotherapy, we follow strict arm care protocols at our institution that prohibits the infusion of chemotherapy into the ipsilateral arm after breast cancer surgery. All patients in our study received chemotherapy via the contralateral arm or central lines. Although we acknowledge that taxane-based chemotherapy may increase the risk of lymphedema, as documented in previous studies, our prediction model was developed specifically as a practical clinical tool for early risk stratification in the outpatient settings. We believe that constructing a simpler prediction model would lead to easier use of our model in daily clinical practice.

Regarding methods for assessing lymphedema, while we acknowledge that there is no universally accepted gold standard, the use of perometry has been shown to provide more reliable and reproducible measurements of limb volume compared to other methods. In our rehabilitation department [3, 4], perometry is the primary diagnostic tool for lymphedema assessment. Arm circumference measurements were performed only in a small subset of patients (less than 5%) who had restricted shoulder mobility due to adhesive capsulitis, in which perometric measurement was not possible.

Dr. Bergmann and Dr. Andrade have also addressed an important issue of tumor size. As for T stage analysis, although it showed significance in the univariate analysis (T3 stage: odds ratio [OR], 7.53; 95% confidence interval [CI], 1.44–35.1; p = 0.009), the multivariate analysis revealed no significant relationships across all stages: T1 stage (OR, 0.87; 95% CI, 0.32–3.02; p = 0.805), T2 stage (OR, 0.82; 95% CI, 0.32–2.89; p = 0.731), T3 stage (OR, 3.22; 95% CI, 0.59–15.57; p = 0.143), and T4 stage (OR, not available [NA]; 95% CI, NA–NA; p = 0.985). Even after regrouping into the early (Tis-T2) and advanced (T3–4) categories, no significant associations were found. We acknowledge that there aren’t many T3–4 cases, but this is because our study only included cases that had sentinel node biopsy. The final prediction model did not include T stage because of its impact on model stability and lack of statistical significance in multivariate analysis.

We thank the editors for their constructive feedback, which helped us strengthen our findings and better understand their limitations.