本篇文獻回顧的目的為何？
我們回顧了與傳統乳房切除術 (去除覆蓋在乳房上的皮膚，包括乳頭和乳暈）相比，皮膚保留乳房切除手術 (SSM) 的手術技術之證據（即去除包括乳房和乳暈（乳頭周圍的皮膚）在內的乳房組織，但保留覆蓋乳房的所有皮膚）。
重點摘要
我們發現 SSM 可能與傳統的乳房切除術在乳癌復發（局部復發）的風險、死於乳癌的機會（整體存活率）或手術後併發症的風險方面沒有什麼不同。評估的手術後併發症包括乳房重建所帶來的損失（由於併發症需要手術切除乳房重建皮瓣或植入物）、皮膚壞死、局部感染、出血（流血）、生活品質和美容效果。本文獻的結果基於 14 項研究，其中大部分可能由於研究設計缺陷而存在偏差。
什麼是皮膚保留乳房切除手術？
傳統的乳癌乳房切除術是一種外科手術，包括切除整個乳房組織、覆蓋乳房的皮膚和乳頭乳暈複合體。此類手術在 20 年後，癌症復發至胸壁（乳房切除術部位）的機率約為 2.3%。為了美觀，人們開始使用皮膚保留乳房切除手術 (SSM) 作為傳統乳房切除術的替代方案。盡可能地保留乳房皮膚、留下最少的乳房組織，並提供更高的心理滿意度和更少的傷害感。在過去的二十年中，SSM 一直用於治療乳癌。
我們想瞭解什麼？
我們想了解 SSM 治療乳癌是否與傳統乳房切除術一樣有效，並評估手術併發症發生率是否不同。
我們做了什麼？
我們檢索了比較 SSM 與其他類型的乳房切除術治療乳癌的研究。我們比較和統整這些研究結果，並且基於研究方法和研究規模等因素，針對相關證據進行可信度評估。
我們發現了什麼？
我們發現了 14 項世代研究 (Cohort study)（長期追蹤的研究），共 12,283 次手術，其中 3,183 人接受了 SSM；9,100 人接受了常規乳房切除術。
接受 SSM 治療或傳統乳房切除術的人可能有類似的情況：
- 手術後癌症復發的可能性（1 項研究）
- 生存（2 項研究）
- 整體併發症的風險（4 項研究）
- 因併發症移除乳房重建皮瓣或植入物的風險（3 項研究）
- 皮膚壞死的風險（4 項研究）
- 感染風險（2 項研究）
- 出血風險（4 項研究）
然而，證據非常不明確。
根據一項研究，SSM 合併即時重建手術與傳統乳房切除手術合併延遲重建，在外觀上沒有差異。一項研究評估了生活品質。該研究結果顯示，接受 SSM 後進行乳房重建的患者與接受乳房切除術但未進行乳房重建的患者，其患者滿意度、社交活動、身體方面和一般狀況相似。
這些證據受到哪些限制？
發現的研究大多是回顧性的。這代表著參與者是在手術多年後被選中，並詢問了他們手術後的經歷，這可能會給研究帶來偏差。在 14 項研究中，有 2 項研究在手術時開始。
接受過傳統乳房切除術的人可能與接受過 SSM 的人不同。大多數研究在分析數據時沒有考慮這些組間差異。
證據更新的時效如何？
本文獻回顧作者搜索截至 2019 年 8 月前已發表之研究報告。
Based on very low-certainty evidence from observational studies, it was not possible to draw definitive conclusions on the effectiveness and safety of SSM for breast cancer treatment. The decision for this technique of breast surgery for treatment of DCIS or invasive breast cancer must be individualized and shared between the physician and the patient while considering the potential risks and benefits of available surgical options.
Skin-sparing mastectomy (SSM) is a surgical technique that aims to maximize skin preservation, facilitate breast reconstruction, and improve cosmetic outcomes. Despite its use in clinical practice, the benefits and harms related to SSM are not well established.
To assess the effectiveness and safety of skin-sparing mastectomy for the treatment of breast cancer.
We searched Cochrane Breast Cancer's Specialized Register, CENTRAL, MEDLINE, Embase, LILACS, the World Health Organization's International Clinical Trials Registry Platform (WHO ICTRP), and ClinicalTrials.gov on 9 August 2019.
Randomized controlled trials (RCTs), quasi-randomized or non-randomized studies (cohort and case-control) comparing SSM to conventional mastectomy for treating ductal carcinoma in situ (DCIS) or invasive breast cancer.
We used standard methodological procedures expected by Cochrane. The primary outcome was overall survival. Secondary outcomes were local recurrence free-survival, adverse events (including overall complications, breast reconstruction loss, skin necrosis, infection and hemorrhage), cosmetic results, and quality of life . We performed a descriptive analysis and meta-analysis of the data.
We found no RCTs or quasi-RCTs. We included two prospective cohort studies and twelve retrospective cohort studies. These studies included 12,211 participants involving 12,283 surgeries (3183 SSM and 9100 conventional mastectomies). It was not possible to perform a meta-analysis for overall survival and local recurrence free-survival due to clinical heterogeneity across studies and a lack of data to calculate hazard ratios (HR).
Based on one study, the evidence suggests that SSM may not reduce overall survival for participants with DCIS tumors (HR 0.41, 95% CI 0.17 to 1.02; P = 0.06; 399 participants; very low-certainty evidence) or for participants with invasive carcinoma (HR 0.81, 95% CI 0.48 to 1.38; P = 0.44; 907 participants; very low-certainty evidence). For local recurrence-free survival, meta-analysis was not possible, due to high risk of bias in nine of the ten studies that measured this outcome. Informal visual examination of effect sizes from nine studies suggested the size of the HR may be similar between groups. Based on one study that adjusted for confounders, SSM may not reduce local recurrence-free survival (HR 0.82, 95% CI 0.47 to 1.42; P = 0.48; 5690 participants; very low-certainty evidence).
The effect of SSM on overall complications is unclear (RR 1.55, 95% CI 0.97 to 2.46; P = 0.07, I 2 = 88%; 4 studies, 677 participants; very low-certainty evidence). Skin-sparing mastectomy may not reduce the risk of breast reconstruction loss (RR 1.79, 95% CI 0.31 to 10.35; P = 0.52; 3 studies, 475 participants; very low-certainty evidence), skin necrosis (RR 1.15, 95% CI 0.62 to 2.12; P = 0.22, I 2 = 33%; 4 studies, 677 participants; very low-certainty evidence), local infection (RR 2.04, 95% CI 0.03 to 142.71; P = 0.74, I 2 = 88%; 2 studies, 371 participants; very low-certainty evidence), nor hemorrhage (RR 1.23, 95% CI 0.47 to 3.27; P = 0.67, I 2 = 0%; 4 studies, 677 participants; very low-certainty evidence). We downgraded the certainty of the evidence due to the risk of bias, imprecision, and inconsistency among the studies. There were no data available on the following outcomes: systemic surgical complications, local complications, explantation of implant/expander, hematoma, seroma, rehospitalization, skin necrosis with revisional surgery, and capsular contracture of the implant.
It was not possible to perform a meta-analysis for cosmetic and quality of life outcomes due to a lack of data. One study performed an evaluation of aesthetic outcome after SSM: 77.7% of participants with immediate breast reconstruction had an overall aesthetic result of excellent or good versus 87% of participants with delayed breast reconstruction.
翻譯者：張綜瑋 (員林基督教醫院，藥師)【本翻譯計畫由臺北醫學大學考科藍臺灣研究中心(Cochrane Taiwan) 及東亞考科藍聯盟 (EACA) 統籌執行。聯絡E-mail:cochranetaiwan@tmu.edu.tw】