Reply

Abstract

Sir: We thank Drs. Longo and Santanelli for their critical evaluation. We are pleased that our article generated such thoughtful inquiry and hope our correspondence will create increased interest in this area of research. Our study population had a mean body mass index of 27, with 54 percent of patients having a body mass index greater than 25. In patients with pendulous hypertrophic breasts, type IV skin-sparing mastectomy using the Wise-pattern technique has documented efficacy regarding breast shape, achieving symmetry and reducing excess skin.1,2 We fully appreciate the special surgical considerations for this patient demographic, but we also recognize that nearly half of our patients had a normal body mass index and thus would not likely benefit from the Wise-pattern technique. Drs. Longo and Santanelli assert that the Wise-pattern skin-sparing mastectomy reduces ischemic complications. However, complication rates approaching 30 percent have been observed with this approach, most often involving skin necrosis at the T-junction.1–3 Our experience mirrors these data, and we do not commonly use the Wise-pattern technique because of a high rate of skin flap necrosis in the triangular flaps. When indicated, a vertical reduction pattern was used, which is associated with significantly reduced ischemic complications of mastectomy skin flaps while retaining the benefits of excess skin removal and excellent aesthetic outcomes.4 Longo and Santanelli state that using Wise-pattern skin-sparing mastectomy would reduce the variability of our results, yet the described modifications to this technique introduce tremendous variability regarding both mechanism of implant coverage and random flap length. Furthermore, standardization of a procedure not indicated in nearly half our patients would contribute little to study homogeneity. The purpose of this study was to identify key risk factors that predict a poor postoperative outcome. Elucidation of the potential influence of skin-sparing mastectomy technique, breast size, or surgical experience on outcomes was beyond the scope of this study and remains of great interest for future investigations. We document that breast radiotherapy had a significant association with implant loss. As we noted in the article, radiotherapy cannot be established as a risk factor for implant loss on the basis of our data alone, given the noted weaknesses in this analysis. We present these data not to establish recommendations but rather to incite further research on the impact of breast radiotherapy on outcomes. In logistic regression, it is suggested that the ratio of events to variables should be at least 10:1 for improved logistic regression modeling, and this ratio in our analysis of implant loss was approximately 4 and 7 for patient- and procedure-related factors, respectively.5 However, studies evaluating small data sets (violating the 10:1 rule) have suggested that this rule is somewhat arbitrary, in part because of its disregard for key parameters such as total sample size.5 All other outcome analyses involved a ratio greater than 10:1. We fully acknowledge the limitations imposed by study design and statistical analysis. However, we feel our data contribute important information to the literature on risk factor identification that will assist in development of criteria permitting preoperative risk stratification. We send our best regards to Drs. Longo and Santanelli. DISCLOSURE The authors have no financial interest to declare in relation to the content of this communication. Edward M. Kobraei, B.S. Brian D. Rinker, M.D. Division of Plastic Surgery University of Kentucky Lexington, Ky.