Reply

Abstract

Sir: We would like to thank Dr. Peeters and Hamdi for their comments regarding our article entitled “Reduced Incidence of Breast Cancer-Related Lymphedema following Mastectomy and Breast Reconstruction versus Mastectomy Alone.”1 The authors present a relevant discussion concerning the underdiagnosis of lymphedema that may be reflected in our study. As described in our Discussion section, one major criticism of lymphedema incidence studies is the criteria used for diagnosis and description of lymphedema. In addition, objective measurements are not standardized and most importantly vary widely.2 As with any retrospective study, there are limitations to how data are documented, as confirmed by the authors. Thus, with further follow-up and better awareness, including improved diagnosis with techniques such as bioimpedance and perometry, more patients may be found to have lymphedema with improved recording of consistent objective measurements. We agree with the authors that information concerning postoperative chemotherapy and lymphedema would contribute to the findings in our article. We did find that neoadjuvant chemotherapy approached significance and, as pointed out by the authors, more recent studies are finding an association with adjuvant chemotherapy and an increased incidence of lymphedema.2,3 To the issue of matching the two study groups, as we know, there are multiple risk factors associated with developing lymphedema. To reduce the confounders in our study, we cross-matched the two groups on three known risk factors: patient age (±5 years), sentinal/axillary lymph node dissection status, and postoperative radiotherapy status. Cross-matching for additional risk factors would reduce the size of the study sample. The more matching variables that are used, the harder it is to find matched pairs, and therefore the more samples we would have excluded from the study, reducing the power of the study. To maintain an adequate sample size for the analysis, we decided to perform cross-matching on only three variables. To consider other risk factors, we applied multivariate analysis. After adjusting for body mass index, the number of positive lymph nodes, and radiotherapy, the conclusion reached by the study was still supported. We described the design and the cross-matching process in the last paragraph of the Discussion. With regard to subgroup analysis, the objective of this study was to compare the incidence of upper limb lymphedema in patients who undergo mastectomy with reconstruction versus mastectomy alone. We discussed the incidence of lymphedema for patients undergoing reconstruction in another article, “Immediate Breast Reconstruction and Lymphedema Incidence,” in which the differences in the lymphedema rates of various reconstruction subgroups were presented and discussed.4 We appreciate the comments by the authors regarding our recent study and look forward to future contributions in the field of lymphedema by Dr. Peeters and Hamdi. We agree that there is a need for improved controlled and prospective trials evaluating both psychosocial and medical outcomes associated with breast cancer–related lymphedema and breast reconstruction. DISCLOSURE None of the authors has a financial interest to declare in relation to the content of this communication. Melissa A. Crosby M.D. Jun Lui M.D., M.S. David W. Chang M.D. Department of Plastic Surgery University of Texas M. D. Anderson Cancer Center Houston, Texas