학회지검색
대한유방검진의학회 학회지
Original Article
성균관대학교 의과대학 삼성서울병원 영상의학과 2성균관대학교 의과대학 삼성서울병원 의생명정보센터 3성균관대학교 의과대학 삼성서울병원 외과4중앙대학교 의과대학 중앙대학교병원 외과5서울대학교 의과대학 서울대학교병원 외과 Effect of Preoperative Magnetic Resonance Imaging on Recurrence Outcomes in Patients with Newly Diagnosed Breast Cancer: A Meta-Analysis Mi-ri Kwon, MD1, Eun Sook Ko, MD, PhD1, Seonwoo Kim, PhD2, Jeong Eon Lee, MD, PhD3, Hee-Chul Shin, MD, PhD4, Dong-Young Noh, MD, PhD5, Boo-Kyung Han, MD, PhD1, Eun Young Ko, MD, PhD1, Ji Soo Choi, MD, PhD1, Ko Woon Park, MD1
1Department of Radiology, Samsung Medical Center, Sungkyunkwan University School of Medicine 2Statistics and Data Center, Samsung Biomedical Research Institute, Samsung Medical Center3Department of Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine 4Department of Surgery, Chung-Ang University College of Medicine, Chung-Ang University Hospital 5Department of Surgery, Seoul National University College of Medicine, Seoul National University Hospital
Background: Several studies have evaluated the relationship between preoperative magnetic resonance imaging (MRI) and recurrence; however, their results are inconsistent.
Purpose: To evaluate the influence of preoperative MRI on recurrence outcome in patients with primary breast cancer.
Materials and Methods: Articles published before April 15, 2017 that included the terms "preoperative breast MRI" were searched for using the PubMed/MEDLINE and EMBASE databases. The eligibility criterion was recurrence outcome according to the use of preoperative MRI with comparative data. The pooled estimate of the hazard ratio (HR) for local or locoregional recurrence was calculated.
Analyses for local, locoregional, contralateral breast recurrence, and distant metastasis were performed. Sources of bias and heterogeneity testing were assessed.
Results: Twelve studies (32,508 patients) met the inclusion criterion. The pooled estimate of HR for local or locoregional recurrence in 10 studies (6,968 patients) in the MRI group was 0.70 (95% confidence interval [CI], 0.41-1.18; P=0.18). HRs for local or locoregional recurrence were heterogeneous between
studies (I2=47.58, P=0.04) and there was significant publication bias. The pooled HR of locoregional recurrence in four studies (2,238 patients) was 0.50 (95% CI, 0.28-0.90; P=0.02) without significant heterogeneity (I2=0.00, P=0.61).
Conclusion: Preoperative breast MRI does not reduce the risk of local and contralateral breast recurrence, while the risk of locoregional recurrence is significantly decreased with limited significance. Thus, it is necessary to establish an indication for preoperative MRI in patients with newly diagnosed breast
cancer.
Key Words: Breast cancer; Meta-analysis; Magnetic resonance ; Prognosis; Recurrence; Preoperative evaluation