Application Of Preoperative Breast MRI On Breast Cancer Patients Undergoing Breast-conserving Surgery

ObjectiveThe morbidity of breast caner have increased significantly in recent years. Breast-conservingsurgery has been used as a standard therapy for early-stage breast cancer at present. Breast-conservingsurgery requires complete removal of the tumor, necessitating histologic negative margins to decreasethe risk of local recurrence. In case of close or tumor-positive resection margins, a re-excision ormastectomy will be performed to achieve an excision with adequate margins. Therefore, accuratepreoperative assessment of tumor size and disease extent is essential for surgical planning.Clinicalexamination, mammography, and ultrasound are the standard techniques to assess tumor size andlocation. However, clinical examination correlates poorly with histopathologic tumor size.Mammography is able to detect clinically occult cancer, including the majority of ductal carcinoma insitu (DCIS), but tends to underestimate the size of invasive breast cancer as well as DCIS, multifocality,and multicentrality, especially in patients with dense fibroglandular tissue. Ultrasound is limited in thedetection of both multifocality and DCIS. Magnetic resonance imaging(MRI) with the highestsensitivity is more and more applied in clinical and can detect mammographically, sonographically, andclinically occult breast cancer, even change the first surgical plan. To evaluate the influence ofpreoperative MRI on the rate of tumor-positive resection margins and surgical plan in breast cancerpatients eligible for breast-conserving surgery, a retrospective study was performed. Furthermore,reasonableness of surgical plan alteration will be evaluated by the final pathological findings Materials and MethodsThe study group consisted of64consecutive patients diagnosed with either breast cancer orductal carcinoma in situ eligible for breast-conserving surgery between March2012and May2013,while73cases between June2010and August2011as the history control group. All the cases werefinally confirmed by histopathology. For the two groups, a first plan for breast-conserving surgery wasmade on the basis of conventional imaging and clinical evaluation. Different from the history controlgroup, the study group underwent preoperative breast MRI as an extra investigation and re-evaluatedbefore final surgical plan was made. To assess the differences in the patient and tumor characteristicsbetween the preoperative MRI group and the historical control group, two-sample Student t tests or theMann-Whitney U test was used for continuous data if appropriate, whereas the chi-square test was usedfor dichotomous data. To compare the differences in the rate of tumor-positive resection marginsbetween the preoperative MRI group and the historical control group, a multivariate logistic regressionanalysis was performed. The covariables included were the group and size of the lesions. Thedifferences between the groups were expressed as odds ratios (ORs) and95%confidence intervals(CIs).To compare the differences in positive resection margins between the two groups for ductal carcinomain situ and invasive malignancy separately, an interaction term (group DCIS vs invasive) was added tothe model. Finally, a binary logistic regression were used to analyze these parameters.Surgical planalteration was recorded in the study group. A significance level of0.05(two-sided) was consideredstatistically significant. Calculations were performed using Stata software (version13.0, Statacorp).ResultsIn the study group,77lesions were found among64patients,average age40.19±0.91years old,and81lesions were found in the history control group,which including73patients, average age39.14±0.92years old. The size of all the lesions as well as the size of the invasive lesions was significantlylarger in the preoperative MRI group compared with the historical control group (p=0.020and p=0.022, respectively). For all other parameters, no significant differences were found between the twogroups. In the preoperative MRI group,fewer patients had tumor-positive resection margins than in the historical control group (9.6%vs24.7%; adjusted OR,0.33;95%CI,0.11-0.94; p=0.04<0.05),and forinvasive carcinoma (P=0.137), whereas for invasive breast caner(P=0.137) there was more decreasingof tumor-positive resection margins than DCIS(P=0.578) between the preoperative MRI group and thehistorical control group.According to the binary logistic regression, both parenchymal density andmalignant calcification could be influence with tumor-positive resection margins.In addition，in12patients (9.4%), a more extensive lumpectomy or a contralateral lumpectomy was performed as a resultof the information obtained from the MRI examination，including5cases additional lesions，6casesrelatively large lesion，1case contralateral breast lesion.Conclusions1. Preoperative breast MRI can reduce the rate of tumor-positive resection margins in breastcancer patients eligible for breast-conserving surgery.2. Both parenchymal density and malignantcalcification are factors which can be influence with tumor-positive resection margins.3. Preoperativebreast MRI can provide more valuable clinical information,even change the first surgical plan.BreastMRI examination is expected to assist clinical treatment,and will play an important role in preoperativeevaluation of breast cancer.