Preoperative Prediction of Sentinel Lymph Node of Breast Cancer Based on MRI
10.3969/j.issn.1005-5185.2024.10.009
- VernacularTitle:MRI术前预测乳腺癌前哨淋巴结转移的价值
- Author:
Yijie SHI
1
;
Qingqing CHEN
;
Jie HE
;
Hongjie HU
Author Information
1. 浙江大学医学院附属邵逸夫医院放射科,浙江 杭州 310000;浙江省医学影像国际科技合作基地,浙江 杭州 310000
- Keywords:
Breast neoplasms;
Sentinel lymph node;
Magnetic resonance imaging;
Pathology,surgical;
Diagnosis,differential;
Forecasting
- From:
Chinese Journal of Medical Imaging
2024;32(10):1021-1026
- CountryChina
- Language:Chinese
-
Abstract:
Purpose To investigate the value of MRI multimodal parameters in predicting sentinel lymph node(SLN)metastasis of breast cancer,and to develop an effective prediction model to reduce the unnecessary biopsy rate of SLN.Materials and Methods Preoperative MRI data of 310 patients with cN0 breast cancer confirmed by operation and pathology in Sir Run Run Shaw Hospital,Zhejiang University School of Medicine from January 2019 to December 2021 were retrospectively analyzed,and all patients were divided into positive group and negative group according to whether SLN metastasized.Imaging features of breast lesions were evaluated independently by two radiologists,differences in parameter between the two groups were compared,independent predictors were screened to build models and evaluate diagnostic efficacy.Results The breast lesions in SLN positive group were located in the upper quadrant(χ2=14.94),non-single(χ2=9.29),circular enhancement(χ2=9.23)and the positive rate of adjacent angiogenesis(χ2=9.91)were higher than those in SLN negative group.The lesions in SLN positive group were larger(Z=-2.97,-2.73),and the early enhancement rate was higher(t=-3.48).The minimum apparent diffusion coefficient and relative apparent diffusion coefficient(minimum apparent diffusion coefficient of lesions/apparent diffusion coefficient of glands,rADC)were lower(Z=-7.33,-10.74),all P<0.05.Logistic regression results showed lesion location(OR=4.17,95%CI 1.86-9.35,P=0.001),early enhancement rate(OR=1.01,95%CI 1.00-1.02,P=0.019)and rADC(OR=54.67,95%CI 23.72-126.02,P<0.001)were independent risk factors for SLN metastasis.The model combining lesion location and rADC had the best predictive performance,with a negative predictive value of 96.5%and a 46.1%reduction in unnecessary biopsies.Conclusion SLN preoperative prediction model(lesion location,rADC)based on conventional MRI characteristic parameters has reliable negative prediction value,which is expected to reduce the unnecessary biopsy of nearly half of the patients with stage cN0 breast cancer.
