From the Department of Radiology, Gangneung Asan Hospital, University of Ulsan College of Medicine, Gangneung, Korea (G.Y.Y.); and Department of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea (G.Y.Y., W.J.C., H.H.K., J.H.C., H.J.S., E.Y.C.).
Radiology. 2020 May;295(2):296-303. doi: 10.1148/radiol.2020191535. Epub 2020 Mar 17.
Background The role of preoperative MRI for predicting surgical outcomes in patients diagnosed with ductal carcinoma in situ (DCIS) remains uncertain. Purpose To investigate the associations between preoperative MRI and surgical outcomes in DCIS confirmed by using US-guided core-needle biopsy (CNB) and to evaluate clinical-pathologic variables associated with a benefit from MRI. Materials and Methods Women with DCIS confirmed by using US-guided CNB between January 2012 and December 2016 were included in this retrospective study. Propensity score matching using 18 confounding covariates was used to create matched groups with MRI and without MRI, and surgical outcomes were compared. Clinical-pathologic variables were evaluated to determine women who benefited from MRI. Results A total of 541 women (mean age ± standard deviation, 50 years ± 10) were evaluated. Among 430 women who underwent MRI, 67 additional lesions (16%) were depicted, with 25 (37%) of the 67 additional lesions being malignant. Fifty-seven (13%) of the 430 women had a change in surgical plan because of their MRI findings; the change was appropriate for 31 (54%) women. In matched groups, the MRI group was associated with lower odds of positive resection margin (odds ratio [OR], 0.39; 95% confidence interval [CI]: 0.16, 0.93; = .03) and repeat surgery (OR, 0.33; 95% CI: 0.12, 0.92; = .03) compared with the non-MRI group. There was no difference in likelihood of initial mastectomy (OR, 1.2; 95% CI: 0.7, 2.0; = .59) and overall mastectomy (OR, 0.93; 95% CI: 0.5, 1.6; = .79). In the MRI group, low nuclear grade (90% [28 of 31] vs 69% [275 of 399]; = .01), progesterone receptor positivity (81% [25 of 31] vs 61% [244 of 399]; = .03), and human epidermal growth factor receptor 2 negativity (90% [28 of 31] vs 68% [270 of 399]; = .01) were associated with a benefit from MRI versus no MRI. Conclusion Preoperative MRI depicted additional malignancy and reduced positive surgical margins and repeat surgery for ductal carcinoma in situ confirmed at US-guided biopsy without a higher mastectomy rate. © RSNA, 2020 See also the editorial by Pinker in this issue.
背景 术前磁共振成像（MRI）在预测导管原位癌（DCIS）患者手术结果中的作用仍不确定。
目的 旨在探讨经超声引导核心针活检（CNB）证实的 DCIS 患者中术前 MRI 与手术结果之间的相关性，并评估与 MRI 获益相关的临床病理变量。
材料与方法 本回顾性研究纳入了 2012 年 1 月至 2016 年 12 月期间经超声引导 CNB 证实为 DCIS 的女性患者。采用 18 个混杂因素的倾向评分匹配，创建了有 MRI 和无 MRI 的匹配组，并比较了手术结果。评估临床病理变量，以确定从 MRI 中获益的患者。
结果 共纳入 541 例女性（平均年龄±标准差，50 岁±10 岁）。在接受 MRI 的 430 例女性中，有 67 个（16%）额外病灶被描绘出来，其中 25 个（37%）为恶性病灶。57 例（13%）女性因 MRI 结果而改变了手术计划，其中 31 例（54%）的改变是恰当的。在匹配组中，与非 MRI 组相比，MRI 组的阳性切缘（比值比[OR]，0.39；95%置信区间[CI]：0.16，0.93；P =.03）和重复手术（OR，0.33；95%CI：0.12，0.92；P =.03）的可能性较低。两组初始乳房切除术（OR，1.2；95%CI：0.7，2.0；P =.59）和总乳房切除术（OR，0.93；95%CI：0.5，1.6；P =.79）的可能性无差异。在 MRI 组中，低核级（90%[28/31]比 69%[275/399]；P =.01）、孕激素受体阳性（81%[25/31]比 61%[244/399]；P =.03）和人表皮生长因子受体 2 阴性（90%[28/31]比 68%[270/399]；P =.01）与 MRI 获益相关，而与无 MRI 获益无关。
结论 术前 MRI 描绘了经超声引导活检证实的 DCIS 患者的额外恶性肿瘤，并降低了阳性手术切缘和重复手术的发生率，而乳房切除术的比例并未增加。
